food-corpus-mixed-D-324f4ab3-c63-20260802

Supposed to	N	Correct	Wrong	Unverified
▸ LOG — log the entry	81	57 (70%)	24 (30%)	0 (0%)
24 handled wrong — click one to jump to its full detail below "I had one cup steel cut oats with water only." OVER-ASK — asked instead of logging (no saved row). "Count twenty cashews." WRITE-TRUTH FAIL — MACROS off for "cashews": logged 3140 kcal vs expected ~160 (§14 15pct/25cal band) "Lunch included four ounces roast turkey." OVER-ASK — asked instead of logging (no saved row). "Ate a half cup dried cranberries." WRITE-TRUTH FAIL — MACROS off for "dried cranberries": logged 246 kcal vs expected ~185 (§14 15pct/25cal band) "Put in one hundred fifty grams skyr yogurt." OVER-ASK — asked instead of logging (no saved row). "Breakfast was three quarters cup ricotta." WRITE-TRUTH FAIL — MACROS off for "ricotta": logged 273 kcal vs expected ~180 (§14 15pct/25cal band) "I had eight ounces 90 percent lean ground beef." OVER-ASK — asked instead of logging (no saved row). "Log three whole wheat pancakes." OVER-ASK — asked instead of logging (no saved row). "I ate one plain bagel." WRITE-TRUTH FAIL — MACROS off for "plain bagel": logged 350 kcal vs expected ~270 (§14 15pct/25cal band) "Record six ounces scallops." WRITE-TRUTH FAIL — MACROS off for "scallops": logged 214 kcal vs expected ~150 (§14 15pct/25cal band) "Log one medium russet potato." OVER-ASK — asked instead of logging (no saved row). "I had one cup butternut squash soup." WRITE-TRUTH FAIL — MACROS off for "butternut squash soup": logged 270 kcal vs expected ~110 (§14 15pct/25cal band) "Put down one cup couscous." OVER-ASK — asked instead of logging (no saved row). "Record five ounces chicken sausage." OVER-ASK — asked instead of logging (no saved row). "Add the same ricotta as yesterday." WRITE-TRUTH FAIL — WRONG/MISSING item "Part Skim Ricotta" — no saved row with matching identity (rows: Ricotta cheese) "Log the same soup I had last night." OVER-ASK — asked instead of logging (no saved row). "Add the same salmon I had earlier." WRITE-TRUTH FAIL — MACROS off for "Baked Salmon Fillet": logged 180 kcal vs expected ~280 (§14 15pct/25cal band) "Snack was one chocolate Chip's Ahoy cookie pack mini." OVER-ASK — asked instead of logging (no saved row). "Log a medium orange and one RXBAR chocolate sea salt." OVER-ASK — asked instead of logging (no saved row). "I had a Kind bar and water." WRITE-TRUTH FAIL — WRONG/MISSING item "a kind bar" — no saved row with matching identity (rows: Cookies & Cream Protein Bar, Water) "Breakfast was avocado toast with egg and chili flakes." OVER-ASK — asked instead of logging (no saved row). "Dinner was a poke bowl with rice, tuna, avocado, and seaweed." OVER-ASK — asked instead of logging (no saved row). "Dinner was butter chicken with rice, naan, and cucumber salad." OVER-ASK — asked instead of logging (no saved row). "I drank a glass of orange juice." OVER-ASK — asked instead of logging (no saved row).
▸ CLARIFY — ask a clarifying question	53	43 (81%)	10 (19%)	0 (0%)
10 handled wrong — click one to jump to its full detail below "Lunch was a tuna melt and an apple." WRONG item vs seeded default/key — IMPLAUSIBLE: 550 kcal is implausibly HIGH for 1 count of Tuna melt sandwich (typical 60-120 kcal [per-serving]) "I ate a pear, string cheese, and pretzels." Logged a BLIND guess — no stated assumption, no correction invited. "Dinner was pork, farro, and Brussels sprouts." Logged a BLIND guess — no stated assumption, no correction invited. "I ate noodles and a side of spinach." WRONG item vs seeded default/key — WRONG/MISSING item "noodles" — no saved row with matching identity (rows: Cooked pasta, Spinach, Cooked pasta, Spinach) "Breakfast was cereal and oat milk." Logged a BLIND guess — no stated assumption, no correction invited. "Log a smoothie with spinach, banana, and whey protein." Logged a BLIND guess — no stated assumption, no correction invited. "I had a grain bowl with quinoa, chickpeas, and tahini." Logged a BLIND guess — no stated assumption, no correction invited. "I had an energy bar." WRONG item vs seeded default/key — WRONG/MISSING item "energy bar" — no saved row with matching identity (rows: Generic protein bar, Generic protein bar, Generic protein bar) "I ate a deli sandwich." Logged a BLIND guess — no stated assumption, no correction invited. "Add a scone." Logged a BLIND guess — no stated assumption, no correction invited.
▸ QUERY — answer the query	18	0 (0%)	0 (0%)	18 (100%)
No errors — all handled correctly.
▸ DELETE — delete the entry	14	5 (36%)	9 (64%)	0 (0%)
9 handled wrong — click one to jump to its full detail below "Delete the banana from earlier." Did not delete (not understood / no removal). "Remove that Quest bar." Did not delete (not understood / no removal). "Delete the oats entry." Did not delete (not understood / no removal). "Take off the eggs." Did not delete (not understood / no removal). "Remove the raspberries." Did not delete (not understood / no removal). "Delete the chicken log." Did not delete (not understood / no removal). "Scratch the almond butter." Did not delete (not understood / no removal). "Remove the milk entry." Did not delete (not understood / no removal). "Take out the wine from last night." Did not delete (not understood / no removal).
▸ UPDATE — update the entry	14	5 (36%)	9 (64%)	0 (0%)
9 handled wrong — click one to jump to its full detail below "Change the sequence banana quantity to two." Reported could-not-confirm (write-confirmation false negative). "Actually make the Quest bar two bars." Reported could-not-confirm (write-confirmation false negative). "Update the oats to one and a half cups." Reported could-not-confirm (write-confirmation false negative). "Change the eggs to four." Reported could-not-confirm (write-confirmation false negative). "Make the raspberries a half cup." Reported could-not-confirm (write-confirmation false negative). "Change almond butter to one tablespoon." Reported could-not-confirm (write-confirmation false negative). "Change the portion to one and a half cups." Reported could-not-confirm (write-confirmation false negative). "Make it a quarter cup instead." Reported could-not-confirm (write-confirmation false negative). "Update that to eight ounces." Reported could-not-confirm (write-confirmation false negative).
▸ CLARIFY_ANSWER — complete the pending action	13	0 (0%)	2 (15%)	11 (85%)
2 handled wrong — click one to jump to its full detail below "Wrong one. The vanilla, not chocolate." Did not complete the pending action. "Not those cashews — the raw ones." Did not complete the pending action.
▸ DECLINE — decline	2	2 (100%)	0 (0%)	0 (0%)
No errors — all handled correctly.
▸ NAVIGATE — navigate	2	1 (50%)	1 (50%)	0 (0%)
1 handled wrong — click one to jump to its full detail below "Open macro goals." Did not navigate — misrouted into a food/workout clarification or dead-ended.
▸ SETTINGS — apply the setting	1	0 (0%)	1 (100%)	0 (0%)
1 handled wrong — click one to jump to its full detail below "Set my protein goal to one hundred sixty grams." Reported could-not-confirm (write-confirmation false negative).
Total	198	113 (67%)	56	29

PASSshould log the entry"Log one large banana for the sequence." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Banana — card not captured.

Why verdict Logged (write-truth verified): Banana — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one large banana for the sequence. Assumed 1 large banana as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one large banana for the sequence. Assumed 1 large banana as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Banana ×1 (1 large banana) 121 cal · 1.5g P · 31.3g C · 0.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:34:52.276Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Banana" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track one Quest birthday cake protein bar." (difficulty —)0.8s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Quest Birthday Cake Protein Bar — card not captured.

Why verdict Logged (write-truth verified): Quest Birthday Cake Protein Bar — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.8s

1 · TTS said Logged Quest Birthday Cake Protein Bar with an estimated macro range.

2 · Card shown Logged Quest Birthday Cake Protein Bar — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Quest Birthday Cake Protein Bar ×1 (one) 136 cal · 19.8g P · 4.2g C · 6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:35:04.222Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"I had one cup steel cut oats with water only." (difficulty —)1.7s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.7s

1 · TTS said I couldn't find Steel Cut Oat's With Water Only Packaged Food in my nutrition database. What exact name is on the package? I did not log it because I'd only be guessing the nutrition. If you prefer, tell me to log a generic packaged food.

2 · Card shown I couldn't find Steel Cut Oat's With Water Only Packaged Food in my nutrition database. What exact name is on the package? I did not log it because I'd only be guessing the nutrition. If you prefer, tell me to log a generic packaged food.

3 · App data rows written No food/workout/list/note data row captured for this utterance.This does not mean "nothing anywhere was stored"; spoken/chat evidence is shown above, and some chat/session records may live outside this write-capture field. It means the report did not capture an app data-row write such as food_log_entries or workout_sets.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:35:17.114Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Add three scrambled eggs for lunch." (difficulty —)0.4s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Egg — card not captured.

Why verdict Logged (write-truth verified): Egg — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said Logged three scrambled eggs for lunch.

2 · Card shown Logged three scrambled eggs for lunch.

3 · App data rows written created food_log_entry: Egg ×1 (3 eggs) 215 cal · 18.9g P · 1g C · 14.3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:35:28.626Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Egg" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Log one cup raspberries as a snack." (difficulty —)0.3s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Raspberries — card not captured.

Why verdict Logged (write-truth verified): Raspberries — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Logged one cup raspberries as a snack. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup raspberries as a snack. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Raspberries ×1 (1 cup) 64 cal · 1.5g P · 14.6g C · 0.9g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:35:40.115Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Raspberries" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Record five ounces baked chicken thigh." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Chicken thigh — card not captured.

Why verdict Logged (write-truth verified): Chicken thigh — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged five ounces baked chicken thigh.

2 · Card shown Logged five ounces baked chicken thigh.

3 · App data rows written created food_log_entry: Chicken thigh ×1 (5 oz) 296 cal · 36.9g P · 0g C · 15.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:35:51.480Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Chicken thigh" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track two tablespoons almond butter." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Almond butter — card not captured.

Why verdict Logged (write-truth verified): Almond butter — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged two tablespoons almond butter.

2 · Card shown Logged two tablespoons almond butter.

3 · App data rows written created food_log_entry: Almond butter ×1 (2 tbsp) 196 cal · 6.7g P · 6.1g C · 17.9g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:36:02.808Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Almond butter" at 90% confidence.
Matched your past food logs to "almond butter" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I drank one cup whole milk." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Whole milk — card not captured.

Why verdict Logged (write-truth verified): Whole milk — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup whole milk. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup whole milk. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Whole milk ×1 (1 cup) 149 cal · 7.8g P · 11.7g C · 8.1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:36:14.202Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Whole milk" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I had one large banana." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Banana — card not captured.

Why verdict Logged (write-truth verified): Banana — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one large banana. Assumed 1 large banana as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one large banana. Assumed 1 large banana as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Banana ×1 (1 large banana) 121 cal · 1.5g P · 31.3g C · 0.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:36:25.502Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Banana" at 90% confidence.
Matched your saved foods and aliases to "Banana" at 83% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Breakfast was three scrambled eggs." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Egg — card not captured.

Why verdict Logged (write-truth verified): Egg — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged three scrambled eggs.

2 · Card shown Logged three scrambled eggs.

3 · App data rows written created food_log_entry: Egg ×1 (3 eggs) 215 cal · 18.9g P · 1g C · 14.3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:36:36.974Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Egg" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Put down five ounces baked chicken thigh." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Chicken thigh — card not captured.

Why verdict Logged (write-truth verified): Chicken thigh — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged five ounces baked chicken thigh.

2 · Card shown Logged five ounces baked chicken thigh.

3 · App data rows written created food_log_entry: Chicken thigh ×1 (5 oz) 296 cal · 36.9g P · 0g C · 15.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:36:48.329Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "five ounces baked chicken thigh" at 90% confidence.
Matched the common-food list to "Chicken thigh" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Record one cup cooked brown rice." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cooked brown rice — card not captured.

Why verdict Logged (write-truth verified): Cooked brown rice — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup cooked brown rice. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup cooked brown rice. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Cooked brown rice ×1 (1 cup) 218 cal · 4.5g P · 45.8g C · 1.6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:36:59.666Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Cooked brown rice" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Snack was a medium pear." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Pear — card not captured.

Why verdict Logged (write-truth verified): Pear — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged a medium pear. Assumed 1 pear as the default serving. Tell me if you want a different amount.

2 · Card shown Logged a medium pear. Assumed 1 pear as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Pear ×1 (1 pear) 101 cal · 0.7g P · 27.1g C · 0.2g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:37:10.974Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Pear" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I ate four ounces baked cod." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cod — card not captured.

Why verdict Logged (write-truth verified): Cod — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged four ounces baked cod.

2 · Card shown Logged four ounces baked cod.

3 · App data rows written created food_log_entry: Cod ×1 (4 oz) 119 cal · 26.1g P · 0g C · 1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:37:22.326Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Cod" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track one cup raspberries." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Raspberries — card not captured.

Why verdict Logged (write-truth verified): Raspberries — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup raspberries. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup raspberries. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Raspberries ×1 (1 cup) 64 cal · 1.5g P · 14.6g C · 0.9g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:37:33.679Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Raspberries" at 90% confidence.
Matched your past food logs to "Raspberries" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Count twenty cashews." (difficulty —)0.9s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — MACROS off for "cashews": logged 3140 kcal vs expected ~160 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "cashews": logged 3140 kcal vs expected ~160 (§14 15pct/25cal band)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.9s

1 · TTS said Got it, I logged Cashews.

2 · Card shown I logged Cashews — let me know if that's not right.

3 · App data rows written created food_log_entry: Cashews ×20 (ounce) 3140 cal · 100g P · 180g C · 240g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:37:45.761Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Lunch included four ounces roast turkey." (difficulty —)4.7s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 4.7s

1 · TTS said What exact food and amount should I use for four ounces roast turkey? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for four ounces roast turkey? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:38:01.821Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Save one cup roasted Brussels sprouts." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Brussels sprouts — card not captured.

Why verdict Logged (write-truth verified): Brussels sprouts — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup roasted brussels sprouts. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup roasted brussels sprouts. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Brussels sprouts ×1 (1 cup) 70 cal · 5.4g P · 14.4g C · 0.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:38:13.224Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Brussels sprouts" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I drank eight ounces apple juice." (difficulty —)0.4s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Apple Juice, 100% — card not captured.

Why verdict Logged (write-truth verified): Apple Juice, 100% — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said Logged Apple Juice, 100% (eight ounces (248 g)).

2 · Card shown Logged Apple Juice, 100% (eight ounces (248 g)).

3 · App data rows written created food_log_entry: Apple Juice, 100% ×1 (eight ounces (248 g)) 119 cal · 0.2g P · 28g C · 0.7g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:38:24.848Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Note three slices Canadian bacon." (difficulty —)0.3s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Canadian bacon — card not captured.

Why verdict Logged (write-truth verified): Canadian bacon — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Logged note three slices canadian bacon.

2 · Card shown Logged note three slices canadian bacon.

3 · App data rows written created food_log_entry: Canadian bacon ×1 (3 slices) 101 cal · 19.5g P · 1.2g C · 1.9g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:38:36.312Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Canadian bacon" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Supper was one cup chickpeas." (difficulty —)0.3s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cooked chickpeas — card not captured.

Why verdict Logged (write-truth verified): Cooked chickpeas — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Logged one cup chickpeas. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup chickpeas. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Cooked chickpeas ×1 (1 cup) 269 cal · 14.6g P · 44.9g C · 4.3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:38:59.443Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Cooked chickpeas" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Enter one medium grapefruit." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Grapefruit — card not captured.

Why verdict Logged (write-truth verified): Grapefruit — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one medium grapefruit. Assumed 1 grapefruit as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one medium grapefruit. Assumed 1 grapefruit as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Grapefruit ×1 (1 grapefruit) 97 cal · 1.8g P · 24.6g C · 0.2g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:39:10.845Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Grapefruit" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Ate a half cup dried cranberries." (difficulty —)0.3s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — MACROS off for "dried cranberries": logged 246 kcal vs expected ~185 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "dried cranberries": logged 246 kcal vs expected ~185 (§14 15pct/25cal band)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Logged ate a half cup dried cranberries. Assumed 0.5 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged ate a half cup dried cranberries. Assumed 0.5 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Dried cranberries ×1 (0.5 cup) 246 cal · 0.2g P · 66.2g C · 0.9g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:39:22.206Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Dried cranberries" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track two tablespoons almond butter." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Almond butter — card not captured.

Why verdict Logged (write-truth verified): Almond butter — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged two tablespoons almond butter.

2 · Card shown Logged two tablespoons almond butter.

3 · App data rows written created food_log_entry: Almond butter ×1 (2 tbsp) 196 cal · 6.7g P · 6.1g C · 17.9g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:39:33.562Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "two tablespoons almond butter" at 90% confidence.
Matched the common-food list to "Almond butter" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Dinner side was one cup asparagus." (difficulty —)0.3s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Asparagus — card not captured.

Why verdict Logged (write-truth verified): Asparagus — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Logged one cup asparagus. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup asparagus. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Asparagus ×1 (1 cup) 40 cal · 4.3g P · 7.4g C · 0.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:39:44.995Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Asparagus" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I had two cups watermelon." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Watermelon — card not captured.

Why verdict Logged (write-truth verified): Watermelon — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged two cups watermelon.

2 · Card shown Logged two cups watermelon.

3 · App data rows written created food_log_entry: Watermelon ×1 (2 cup) 91 cal · 1.8g P · 23.1g C · 0.6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:39:56.306Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Watermelon" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Put in one hundred fifty grams skyr yogurt." (difficulty —)0.3s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said I found Plain Greek yogurt, but I also heard hundred fifty skyr yogurt. Should I log those separately too?

2 · Card shown I found Plain Greek yogurt, but I also heard hundred fifty skyr yogurt. Should I log those separately too?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:40:08.122Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Plain Greek yogurt" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Log one cup whole milk." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Whole milk — card not captured.

Why verdict Logged (write-truth verified): Whole milk — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup whole milk. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup whole milk. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Whole milk ×1 (1 cup) 149 cal · 7.8g P · 11.7g C · 8.1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:40:19.438Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "one cup whole milk" at 90% confidence.
Matched the common-food list to "Whole milk" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Breakfast was three quarters cup ricotta." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — MACROS off for "ricotta": logged 273 kcal vs expected ~180 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "ricotta": logged 273 kcal vs expected ~180 (§14 15pct/25cal band)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged three quarters cup ricotta.

2 · Card shown Logged three quarters cup ricotta.

3 · App data rows written created food_log_entry: Ricotta cheese ×1 (0.75 cup) 273 cal · 17.7g P · 11.1g C · 17.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:40:30.774Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Ricotta cheese" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Count two small flour tortillas." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Flour tortilla — card not captured.

Why verdict Logged (write-truth verified): Flour tortilla — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged two small flour tortillas.

2 · Card shown Logged two small flour tortillas.

3 · App data rows written created food_log_entry: Flour tortilla ×1 (2 small flour tortillas) 149 cal · 4.1g P · 24g C · 3.8g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:40:42.128Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Flour tortilla" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"I had eight ounces 90 percent lean ground beef." (difficulty —)5.0s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 5.0s

1 · TTS said Which one did you mean? 1. Lean Ground Beef — BUTCHER'S BRAND, 2. Lean Ground Beef — FRESHLAST PACK, 3. Lean Ground Beef — The Kroger Co., 4. Lean Ground Beef — FPL Food LLC

2 · Card shown Which one did you mean? 1. Lean Ground Beef — BUTCHER'S BRAND, 2. Lean Ground Beef — FRESHLAST PACK, 3. Lean Ground Beef — The Kroger Co., 4. Lean Ground Beef — FPL Food LLC

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:40:58.281Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Add one cup pinto beans." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Pinto beans — card not captured.

Why verdict Logged (write-truth verified): Pinto beans — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup pinto beans. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup pinto beans. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Pinto beans ×1 (1 cup) 245 cal · 15.4g P · 44.5g C · 1.2g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:41:09.614Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Pinto beans" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track one whole avocado." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Avocado — card not captured.

Why verdict Logged (write-truth verified): Avocado — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one whole avocado. Assumed 1 avocado as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one whole avocado. Assumed 1 avocado as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Avocado ×1 (1 avocado) 240 cal · 3g P · 12.8g C · 22g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:41:20.930Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Avocado" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Log three whole wheat pancakes." (difficulty —)1.4s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.4s

1 · TTS said Which one did you mean? 1. Whole Wheat Pancakes — Echo Lake Foods, 2. Whole Wheat Pancakes — 365 EVERYDAY VALUE

2 · Card shown Which one did you mean? 1. Whole Wheat Pancakes — Echo Lake Foods, 2. Whole Wheat Pancakes — 365 EVERYDAY VALUE

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:41:33.461Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"I ate one plain bagel." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — MACROS off for "plain bagel": logged 350 kcal vs expected ~270 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "plain bagel": logged 350 kcal vs expected ~270 (§14 15pct/25cal band)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one plain bagel. Assumed a fresh/deli-size bagel; say packaged or Lender's if it was a bread-aisle bagel Assumed 1 plain bagel as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one plain bagel. Assumed a fresh/deli-size bagel; say packaged or Lender's if it was a bread-aisle bagel Assumed 1 plain bagel as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Plain bagel ×1 (1 plain bagel) 350 cal · 14g P · 67.2g C · 2.1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:41:44.800Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Plain bagel" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Record six ounces scallops." (difficulty —)0.7s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — MACROS off for "scallops": logged 214 kcal vs expected ~150 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "scallops": logged 214 kcal vs expected ~150 (§14 15pct/25cal band)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.7s

1 · TTS said Got it, I logged Scallops, Baked or Broiled (six ounces (170.1 g)).

2 · Card shown I logged Scallops, Baked or Broiled (six ounces (170.1 g)) — let me know if that's not right.

3 · App data rows written created food_log_entry: Scallops, Baked or Broiled ×1 (six ounces (170.1 g)) 214 cal · 25.5g P · 6.8g C · 8.7g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:41:56.684Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Snack was two Babybel cheeses." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Babybel — card not captured.

Why verdict Logged (write-truth verified): Babybel — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged two babybel cheeses.

2 · Card shown Logged two babybel cheeses.

3 · App data rows written created food_log_entry: Babybel ×1 (2 babybel) 140 cal · 10g P · 0g C · 11g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:42:08.081Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Babybel" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Add two teaspoons butter." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Butter — card not captured.

Why verdict Logged (write-truth verified): Butter — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged two teaspoons butter.

2 · Card shown Logged two teaspoons butter.

3 · App data rows written created food_log_entry: Butter ×1 (2 tsp) 68 cal · 0.1g P · 0g C · 7.7g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:42:19.471Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Butter" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I had one cup farro." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cooked farro — card not captured.

Why verdict Logged (write-truth verified): Cooked farro — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup farro. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup farro. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Cooked farro ×1 (1 cup) 240 cal · 8.6g P · 50.7g C · 1.6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:42:30.815Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Cooked farro" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Log one medium russet potato." (difficulty —)3.5s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.5s

1 · TTS said Was the potato plain, or did it have toppings like butter, sour cream, or cheese? I did not log it yet because add-ons can change the nutrition.

2 · Card shown Was the potato plain, or did it have toppings like butter, sour cream, or cheese? I did not log it yet because add-ons can change the nutrition.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:42:45.568Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track ten ounces green tea." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Green tea — card not captured.

Why verdict Logged (write-truth verified): Green tea — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged ten ounces green tea.

2 · Card shown Logged ten ounces green tea.

3 · App data rows written created food_log_entry: Green tea ×1 (10 oz) 3 cal · 0g P · 0g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:43:08.664Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Green tea" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Count one cup cherry tomatoes." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cherry tomatoes — card not captured.

Why verdict Logged (write-truth verified): Cherry tomatoes — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup cherry tomatoes. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup cherry tomatoes. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Cherry tomatoes ×1 (1 cup) 27 cal · 1.3g P · 5.8g C · 0.3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:43:20.016Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Cherry tomatoes" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Dinner was six ounces pork tenderloin." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Pork tenderloin — card not captured.

Why verdict Logged (write-truth verified): Pork tenderloin — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged six ounces pork tenderloin.

2 · Card shown Logged six ounces pork tenderloin.

3 · App data rows written created food_log_entry: Pork tenderloin ×1 (6 oz) 243 cal · 44.2g P · 0g C · 6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:43:31.312Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Pork tenderloin" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"I had one cup butternut squash soup." (difficulty —)1.4s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — MACROS off for "butternut squash soup": logged 270 kcal vs expected ~110 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "butternut squash soup": logged 270 kcal vs expected ~110 (§14 15pct/25cal band)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.4s

1 · TTS said Logged The Cheesecake Factory Cup of Butternut Squash Soup with an estimated macro range.

2 · Card shown Logged The Cheesecake Factory Cup of Butternut Squash Soup — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: The Cheesecake Factory Cup of Butternut Squash Soup ×1 (1 cup) 270 cal · 3g P · 21g C · 21g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:43:43.876Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Add three rice cakes." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Plain rice cakes — card not captured.

Why verdict Logged (write-truth verified): Plain rice cakes — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged three rice cakes.

2 · Card shown Logged three rice cakes.

3 · App data rows written created food_log_entry: Plain rice cakes ×1 (3 rice cakes) 104 cal · 2.2g P · 22g C · 0.8g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:43:55.194Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Plain rice cakes" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Log one cup blackberries." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Blackberries — card not captured.

Why verdict Logged (write-truth verified): Blackberries — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one cup blackberries. Assumed 1 cup as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one cup blackberries. Assumed 1 cup as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Blackberries ×1 (1 cup) 62 cal · 2g P · 14.7g C · 0.7g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:44:06.515Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Blackberries" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track two tablespoons guacamole." (difficulty —)0.7s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Guacamole — card not captured.

Why verdict Logged (write-truth verified): Guacamole — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.7s

1 · TTS said Got it, I logged Guacamole.

2 · Card shown I logged Guacamole — let me know if that's not right.

3 · App data rows written created food_log_entry: Guacamole ×1 (tablespoon) 50 cal · 1g P · 3g C · 4.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:44:18.520Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I ate one medium nectarine." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Nectarine — card not captured.

Why verdict Logged (write-truth verified): Nectarine — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged one medium nectarine. Assumed 1 nectarine as the default serving. Tell me if you want a different amount.

2 · Card shown Logged one medium nectarine. Assumed 1 nectarine as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Nectarine ×1 (1 nectarine) 62 cal · 1.5g P · 14.8g C · 0.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:44:29.812Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Nectarine" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Put down one cup couscous." (difficulty —)0.8s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.8s

1 · TTS said I couldn't convert one cup of Couscous, Dry from the serving information I found. What does the label say for grams or calories? I did not log it yet.

2 · Card shown I couldn't convert one cup of Couscous, Dry from the serving information I found. What does the label say for grams or calories? I did not log it yet.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:44:41.763Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Record five ounces chicken sausage." (difficulty —)1.5s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.5s

1 · TTS said Which type of chicken sausage was it? There are several with different nutrition.

2 · Card shown Which type of chicken sausage was it? There are several with different nutrition.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:44:54.446Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Add my usual RXBAR." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): RXBAR Blueberry — card not captured.

Why verdict Logged (write-truth verified): RXBAR Blueberry — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged RXBAR RXBAR Blueberry. Using your saved "my usual rxbar". Tell me if that is wrong.

2 · Card shown Logged RXBAR RXBAR Blueberry. Using your saved "my usual rxbar". Tell me if that is wrong.

3 · App data rows written created food_log_entry: RXBAR Blueberry ×1 (bar) 210 cal · 12g P · 25g C · 8g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: RXBAR RXBAR Blueberry

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-rxbar",
    "name": "RXBAR Blueberry",
    "brand": "RXBAR",
    "source": "manual",
    "servingLabel": "bar",
    "servingQuantity": 1,
    "servingUnit": "bar",
    "macrosPerServing": {
      "calories": 210,
      "proteinGrams": 12,
      "fatGrams": 8,
      "carbsGrams": 25
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual RXBAR",
  "foodItemId": "seed-d-rxbar",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "bar",
    "displayLabel": "my usual RXBAR",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:45:05.717Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual rxbar" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I had my usual skyr." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Plain Skyr — card not captured.

Why verdict Logged (write-truth verified): Plain Skyr — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Codex Test Dairy Plain Skyr. Using your saved "my usual skyr". Tell me if that is wrong.

2 · Card shown Logged Codex Test Dairy Plain Skyr. Using your saved "my usual skyr". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Plain Skyr ×1 (cup) 110 cal · 19g P · 7g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Dairy Plain Skyr

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-skyr",
    "name": "Plain Skyr",
    "brand": "Codex Test Dairy",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 110,
      "proteinGrams": 19,
      "fatGrams": 0,
      "carbsGrams": 7
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual skyr",
  "foodItemId": "seed-d-skyr",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual skyr",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:45:17.026Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual skyr" at 96% confidence.
Matched the common-food list to "Skyr" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Lunch was my usual chicken salad." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Chicken Salad Scoop — card not captured.

Why verdict Logged (write-truth verified): Chicken Salad Scoop — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Codex Test Deli Chicken Salad Scoop. Using your saved "my usual chicken salad". Tell me if that is wrong.

2 · Card shown Logged Codex Test Deli Chicken Salad Scoop. Using your saved "my usual chicken salad". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Chicken Salad Scoop ×1 (scoop) 280 cal · 22g P · 6g C · 18g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Deli Chicken Salad Scoop

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-chix-salad",
    "name": "Chicken Salad Scoop",
    "brand": "Codex Test Deli",
    "source": "manual",
    "servingLabel": "scoop",
    "servingQuantity": 1,
    "servingUnit": "scoop",
    "macrosPerServing": {
      "calories": 280,
      "proteinGrams": 22,
      "fatGrams": 18,
      "carbsGrams": 6
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual chicken salad",
  "foodItemId": "seed-d-chix-salad",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "scoop",
    "displayLabel": "my usual chicken salad",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:45:28.422Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual chicken salad" at 96% confidence.
Matched the common-food list to "Chicken salad with mayonnaise" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Dinner side was my usual quinoa." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cooked Quinoa — card not captured.

Why verdict Logged (write-truth verified): Cooked Quinoa — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Cooked Quinoa. Using your saved "my usual quinoa". Tell me if that is wrong.

2 · Card shown Logged Cooked Quinoa. Using your saved "my usual quinoa". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Cooked Quinoa ×1 (cup) 222 cal · 8g P · 39g C · 3.6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Cooked Quinoa

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-quinoa",
    "name": "Cooked Quinoa",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 222,
      "proteinGrams": 8,
      "fatGrams": 3.6,
      "carbsGrams": 39
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual quinoa",
  "foodItemId": "seed-d-quinoa",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual quinoa",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:45:39.662Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual quinoa" at 96% confidence.
Matched the common-food list to "Cooked quinoa" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track my usual cashews." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cashews — card not captured.

Why verdict Logged (write-truth verified): Cashews — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Cashews. Using your saved "my usual cashews". Tell me if that is wrong.

2 · Card shown Logged Cashews. Using your saved "my usual cashews". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Cashews ×1 (ounce) 157 cal · 5g P · 9g C · 12g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Cashews

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-cashews",
    "name": "Cashews",
    "brand": null,
    "source": "manual",
    "servingLabel": "ounce",
    "servingQuantity": 1,
    "servingUnit": "oz",
    "macrosPerServing": {
      "calories": 157,
      "proteinGrams": 5,
      "fatGrams": 12,
      "carbsGrams": 9
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual cashews",
  "foodItemId": "seed-d-cashews",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "ounce",
    "displayLabel": "my usual cashews",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:45:50.912Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual cashews" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Log my default oat milk." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Oat Milk — card not captured.

Why verdict Logged (write-truth verified): Oat Milk — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Oat Milk. Using your saved "my default oat milk". Tell me if that is wrong.

2 · Card shown Logged Oat Milk. Using your saved "my default oat milk". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Oat Milk ×1 (cup) 120 cal · 3g P · 16g C · 5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Oat Milk

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-oatmilk",
    "name": "Oat Milk",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 120,
      "proteinGrams": 3,
      "fatGrams": 5,
      "carbsGrams": 16
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my default oat milk",
  "foodItemId": "seed-d-oatmilk",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my default oat milk",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:46:02.178Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my default oat milk" at 96% confidence.
Matched the common-food list to "Milk" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Add my usual evening cocoa." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Hot Cocoa — card not captured.

Why verdict Logged (write-truth verified): Hot Cocoa — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Codex Test Cafe Hot Cocoa. Using your saved "my usual evening cocoa". Tell me if that is wrong.

2 · Card shown Logged Codex Test Cafe Hot Cocoa. Using your saved "my usual evening cocoa". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Hot Cocoa ×1 (cup) 140 cal · 4g P · 24g C · 3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Cafe Hot Cocoa

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-cocoa",
    "name": "Hot Cocoa",
    "brand": "Codex Test Cafe",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 140,
      "proteinGrams": 4,
      "fatGrams": 3,
      "carbsGrams": 24
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual evening cocoa",
  "foodItemId": "seed-d-cocoa",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual evening cocoa",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:46:13.508Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual evening cocoa" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had my normal steel cut oats." (difficulty —)0.1s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Steel Cut Oats. Using your saved "my normal steel cut oats". Tell me if that is wrong.

2 · Card shown Logged Steel Cut Oats. Using your saved "my normal steel cut oats". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Steel Cut Oats ×1 (cup) 170 cal · 7g P · 29g C · 3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Steel Cut Oats

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-oats",
    "name": "Steel Cut Oats",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 170,
      "proteinGrams": 7,
      "fatGrams": 3,
      "carbsGrams": 29
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my normal steel cut oats",
  "foodItemId": "seed-d-oats",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my normal steel cut oats",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:46:24.894Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my normal steel cut oats" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track my go-to latte." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Caffe Latte — card not captured.

Why verdict Logged (write-truth verified): Caffe Latte — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Codex Test Cafe Caffe Latte. Using your saved "my go to latte". Tell me if that is wrong.

2 · Card shown Logged Codex Test Cafe Caffe Latte. Using your saved "my go to latte". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Caffe Latte ×1 (cup) 150 cal · 8g P · 14g C · 6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Cafe Caffe Latte

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-latte",
    "name": "Caffe Latte",
    "brand": "Codex Test Cafe",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 8,
      "fatGrams": 6,
      "carbsGrams": 14
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my go-to latte",
  "foodItemId": "seed-d-latte",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my go-to latte",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:46:36.202Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my go to latte" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Log my usual bedtime tea." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Chamomile Tea — card not captured.

Why verdict Logged (write-truth verified): Chamomile Tea — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Chamomile Tea. Using your saved "my usual bedtime tea". Tell me if that is wrong.

2 · Card shown Logged Chamomile Tea. Using your saved "my usual bedtime tea". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Chamomile Tea ×1 (cup) 0 cal · 0g P · 0g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Chamomile Tea

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-tea",
    "name": "Chamomile Tea",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 0,
      "proteinGrams": 0,
      "fatGrams": 0,
      "carbsGrams": 0
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual bedtime tea",
  "foodItemId": "seed-d-tea",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual bedtime tea",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T21:46:47.453Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual bedtime tea" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Add the same ricotta as yesterday." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — WRONG/MISSING item "Part Skim Ricotta" — no saved row with matching identity (rows: Ricotta cheese)

Why verdict WRITE-TRUTH FAIL — WRONG/MISSING item "Part Skim Ricotta" — no saved row with matching identity (rows: Ricotta cheese)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged Ricotta cheese. Using your recent Ricotta cheese history. Tell me if that is wrong.

2 · Card shown Logged Ricotta cheese. Using your recent Ricotta cheese history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Ricotta cheese ×1 (0.75 cup) 273 cal · 17.7g P · 11.1g C · 17.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Dairy Part Skim Ricotta

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-ricotta",
    "name": "Part Skim Ricotta",
    "brand": "Codex Test Dairy",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 14,
      "fatGrams": 12,
      "carbsGrams": 6
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 0.75 for seed-d-ricotta at 2026-07-17T13:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-ricotta",
  "quantity": 0.75,
  "loggedAt": "2026-07-17T13:00:00.000Z"
}

snapshot captured 2026-08-02T21:47:11.254Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Ricotta cheese" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Same granola as yesterday." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Honey Granola — card not captured.

Why verdict Logged (write-truth verified): Honey Granola — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged Codex Test Pantry Honey Granola. Using your recent Codex Test Pantry Honey Granola history. Tell me if that is wrong.

2 · Card shown Logged Codex Test Pantry Honey Granola. Using your recent Codex Test Pantry Honey Granola history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Honey Granola ×1 (cup) 240 cal · 6g P · 38g C · 8g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Pantry Honey Granola

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-yest-granola",
    "name": "Honey Granola",
    "brand": "Codex Test Pantry",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 240,
      "proteinGrams": 6,
      "fatGrams": 8,
      "carbsGrams": 38
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 0.5 for seed-d-yest-granola at 2026-07-17T08:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-yest-granola",
  "quantity": 0.5,
  "loggedAt": "2026-07-17T08:00:00.000Z"
}

snapshot captured 2026-08-02T21:47:22.666Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Granola" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Log the same soup I had last night." (difficulty —)0.5s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.5s

1 · TTS said I heard you want the same the same soup I had last night as before, but I do not have a matching saved log yet. Which exact food and amount should I log?

2 · Card shown I heard you want the same the same soup I had last night as before, but I do not have a matching saved log yet. Which exact food and amount should I log?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Kitchen Lentil Soup

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-soup",
    "name": "Lentil Soup",
    "brand": "Codex Test Kitchen",
    "source": "manual",
    "servingLabel": "bowl",
    "servingQuantity": 1,
    "servingUnit": "bowl",
    "macrosPerServing": {
      "calories": 220,
      "proteinGrams": 12,
      "fatGrams": 4,
      "carbsGrams": 34
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-soup at 2026-07-17T19:30:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-soup",
  "quantity": 1,
  "loggedAt": "2026-07-17T19:30:00.000Z"
}

snapshot captured 2026-08-02T21:47:34.306Z

6 · Why this food The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Same bagel as this morning." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Everything Bagel — card not captured.

Why verdict Logged (write-truth verified): Everything Bagel — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged Everything Bagel. Using your recent Everything Bagel history. Tell me if that is wrong.

2 · Card shown Logged Everything Bagel. Using your recent Everything Bagel history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Everything Bagel ×1 (bagel) 280 cal · 10g P · 54g C · 2g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Everything Bagel

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-hist-bagel",
    "name": "Everything Bagel",
    "brand": null,
    "source": "manual",
    "servingLabel": "bagel",
    "servingQuantity": 1,
    "servingUnit": "bagel",
    "macrosPerServing": {
      "calories": 280,
      "proteinGrams": 10,
      "fatGrams": 2,
      "carbsGrams": 54
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-hist-bagel at 2026-07-18T07:30:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-hist-bagel",
  "quantity": 1,
  "loggedAt": "2026-07-18T07:30:00.000Z"
}

snapshot captured 2026-08-02T21:47:45.726Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Plain bagel" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Add the same salmon I had earlier." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — MACROS off for "Baked Salmon Fillet": logged 180 kcal vs expected ~280 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "Baked Salmon Fillet": logged 180 kcal vs expected ~280 (§14 15pct/25cal band)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged salmon. Using your recent salmon history. Tell me if that is wrong.

2 · Card shown Logged salmon. Using your recent salmon history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: salmon ×1 (serving) 180 cal · 8g P · 18g C · 6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Baked Salmon Fillet

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-hist-salmon",
    "name": "Baked Salmon Fillet",
    "brand": null,
    "source": "manual",
    "servingLabel": "fillet",
    "servingQuantity": 1,
    "servingUnit": "fillet",
    "macrosPerServing": {
      "calories": 280,
      "proteinGrams": 34,
      "fatGrams": 14,
      "carbsGrams": 0
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-hist-salmon at 2026-07-18T12:30:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-hist-salmon",
  "quantity": 1,
  "loggedAt": "2026-07-18T12:30:00.000Z"
}

snapshot captured 2026-08-02T21:47:57.074Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Salmon" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Same smoothie as yesterday." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Green Protein Smoothie — card not captured.

Why verdict Logged (write-truth verified): Green Protein Smoothie — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged Codex Test Recipe Green Protein Smoothie. Using your recent Codex Test Recipe Green Protein Smoothie history. Tell me if that is wrong.

2 · Card shown Logged Codex Test Recipe Green Protein Smoothie. Using your recent Codex Test Recipe Green Protein Smoothie history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Green Protein Smoothie ×1 (smoothie) 260 cal · 24g P · 30g C · 5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Recipe Green Protein Smoothie

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-smoothie",
    "name": "Green Protein Smoothie",
    "brand": "Codex Test Recipe",
    "source": "manual",
    "servingLabel": "smoothie",
    "servingQuantity": 1,
    "servingUnit": "smoothie",
    "macrosPerServing": {
      "calories": 260,
      "proteinGrams": 24,
      "fatGrams": 5,
      "carbsGrams": 30
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-smoothie at 2026-07-17T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-smoothie",
  "quantity": 1,
  "loggedAt": "2026-07-17T10:00:00.000Z"
}

snapshot captured 2026-08-02T21:48:08.433Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Log the tofu I had last time." (difficulty —)0.1s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Firm Tofu — card not captured.

Why verdict Logged (write-truth verified): Firm Tofu — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged Firm Tofu. Using your saved "Firm Tofu". Tell me if that is wrong.

2 · Card shown Logged Firm Tofu. Using your saved "Firm Tofu". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Firm Tofu ×1 (serving) 140 cal · 15g P · 3g C · 8g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Firm Tofu

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-hist-tofu",
    "name": "Firm Tofu",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 140,
      "proteinGrams": 15,
      "fatGrams": 8,
      "carbsGrams": 3
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-hist-tofu at 2026-07-16T18:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-hist-tofu",
  "quantity": 1,
  "loggedAt": "2026-07-16T18:00:00.000Z"
}

snapshot captured 2026-08-02T21:48:19.697Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Firm tofu" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Same toast as earlier today." (difficulty —)0.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Sourdough Toast — card not captured.

Why verdict Logged (write-truth verified): Sourdough Toast — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Logged Sourdough Toast. Using your recent Sourdough Toast history. Tell me if that is wrong.

2 · Card shown Logged Sourdough Toast. Using your recent Sourdough Toast history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Sourdough Toast ×1 (slice) 90 cal · 3g P · 17g C · 1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Sourdough Toast

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-toast",
    "name": "Sourdough Toast",
    "brand": null,
    "source": "manual",
    "servingLabel": "slice",
    "servingQuantity": 1,
    "servingUnit": "slice",
    "macrosPerServing": {
      "calories": 90,
      "proteinGrams": 3,
      "fatGrams": 1,
      "carbsGrams": 17
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 2 for seed-d-toast at 2026-07-18T07:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-toast",
  "quantity": 2,
  "loggedAt": "2026-07-18T07:00:00.000Z"
}

snapshot captured 2026-08-02T21:48:31.042Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Toast" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Snack was one chocolate Chip's Ahoy cookie pack mini." (difficulty —)2.6s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.6s

1 · TTS said What food did you have?

2 · Card shown What food did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:48:44.781Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Finished one Fairlife Core Power vanilla shake." (difficulty —)1.2s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Fairlife Core Power Elite Vanilla Protein Shake — card not captured.

Why verdict Logged (write-truth verified): Fairlife Core Power Elite Vanilla Protein Shake — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.2s

1 · TTS said Logged Fairlife Core Power Elite Vanilla Protein Shake with an estimated macro range.

2 · Card shown Logged Fairlife Core Power Elite Vanilla Protein Shake — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Fairlife Core Power Elite Vanilla Protein Shake ×1 (one) 230 cal · 42g P · 8g C · 3.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:48:57.215Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Track one Chobani zero sugar strawberry cup." (difficulty —)0.6s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Chobani Greek Yogurt, Strawberry, Zero Sugar — card not captured.

Why verdict Logged (write-truth verified): Chobani Greek Yogurt, Strawberry, Zero Sugar — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.6s

1 · TTS said Logged Chobani Greek Yogurt, Strawberry, Zero Sugar.

2 · Card shown Logged Chobani Greek Yogurt, Strawberry, Zero Sugar.

3 · App data rows written created food_log_entry: Chobani Greek Yogurt, Strawberry, Zero Sugar ×1 (one) 60 cal · 12g P · 5g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:49:08.999Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Save one Quest cookies and cream protein bar." (difficulty —)0.5s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cookies & Cream Protein Bar — card not captured.

Why verdict Logged (write-truth verified): Cookies & Cream Protein Bar — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.5s

1 · TTS said Logged Cookies & Cream Protein Bar — Quest with an estimated macro range.

2 · Card shown Logged Cookies & Cream Protein Bar — Quest — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Cookies & Cream Protein Bar ×1 (1 BAR) 200 cal · 21g P · 21g C · 8g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:49:20.685Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I ate one KIND peanut butter dark chocolate bar." (difficulty —)0.5s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): KIND Peanut Butter Dark Chocolate Bar — card not captured.

Why verdict Logged (write-truth verified): KIND Peanut Butter Dark Chocolate Bar — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.5s

1 · TTS said Got it, I logged KIND Peanut Butter Dark Chocolate Bar.

2 · Card shown I logged KIND Peanut Butter Dark Chocolate Bar — let me know if that's not right.

3 · App data rows written created food_log_entry: KIND Peanut Butter Dark Chocolate Bar ×1 (1 kind peanut butter dark chocolate bar) 200 cal · 7g P · 17g C · 14g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:49:32.408Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Log a medium orange and one RXBAR chocolate sea salt." (difficulty —)1.6s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.6s

1 · TTS said I need to resolve one RXBAR chocolate sea salt before I log this meal. What should I use for one RXBAR chocolate sea salt?

2 · Card shown I need to resolve one RXBAR chocolate sea salt before I log this meal. What should I use for one RXBAR chocolate sea salt?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:49:45.207Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had steel cut oats and green tea." (difficulty —)0.5s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.5s

1 · TTS said Logged Steel Cut Oats and Green tea. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

2 · Card shown Logged Steel Cut Oats and Green tea. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

3 · App data rows written created food_log_entry: Steel Cut Oats ×1 (cup) 170 cal · 7g P · 29g C · 3g F
created food_log_entry: Green tea ×1 (240 g) 2 cal · 0g P · 0g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:49:56.801Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Breakfast was scrambled eggs and a bagel." (difficulty —)0.3s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said How much should I log for Egg? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Egg? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:50:08.417Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Plain bagel" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"Lunch was a tuna melt and an apple." (difficulty —)0.3s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: WRONG item vs seeded default/key — IMPLAUSIBLE: 550 kcal is implausibly HIGH for 1 count of Tuna melt sandwich (typical 60-120 kcal [per-serving])

Why verdict WRONG item vs seeded default/key — IMPLAUSIBLE: 550 kcal is implausibly HIGH for 1 count of Tuna melt sandwich (typical 60-120 kcal [per-serving])

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Logged Tuna melt sandwich and an apple. Assumed 1 apple as the default serving. Tell me if you want a different amount. Tell me if that is not right.

2 · Card shown Logged Tuna melt sandwich and an apple. Assumed 1 apple as the default serving. Tell me if you want a different amount. Tell me if that is not right.

3 · App data rows written created food_log_entry: Tuna melt sandwich ×1 (1 tuna melt) 550 cal · 30.8g P · 44g C · 26.4g F
created food_log_entry: Apple ×1 (1 apple) 95 cal · 0.5g P · 25.5g C · 0.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:50:19.945Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Snack was cashews and a pear." (difficulty —)2.2s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.2s

1 · TTS said I need to resolve cashews before I log this meal. What should I use for cashews?

2 · Card shown I need to resolve cashews before I log this meal. What should I use for cashews?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:50:33.313Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Dinner was cod and asparagus." (difficulty —)0.3s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said How much should I log for Cod and Asparagus? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Cod and Asparagus? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:50:44.722Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Asparagus" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I ate skyr and raspberries." (difficulty —)0.3s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said How much should I log for Skyr and Raspberries? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Skyr and Raspberries? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:50:56.132Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Raspberries" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Track turkey and quinoa." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said How much should I log for Turkey breast and Cooked quinoa? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Turkey breast and Cooked quinoa? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:51:19.327Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Cooked quinoa" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add ricotta and blackberries." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said How much should I log for Ricotta cheese and Blackberries? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Ricotta cheese and Blackberries? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:51:30.874Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Blackberries" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"I had a Kind bar and water." (difficulty —)0.8s

Verdict Expected LOG — should log the entry. FAIL: WRITE-TRUTH FAIL — WRONG/MISSING item "a kind bar" — no saved row with matching identity (rows: Cookies & Cream Protein Bar, Water)

Why verdict WRITE-TRUTH FAIL — WRONG/MISSING item "a kind bar" — no saved row with matching identity (rows: Cookies & Cream Protein Bar, Water)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.8s

1 · TTS said Logged QUEST Cookies & Cream Protein Bar and Water. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

2 · Card shown Logged QUEST Cookies & Cream Protein Bar and Water. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

3 · App data rows written created food_log_entry: Cookies & Cream Protein Bar ×1 (1 BAR) 200 cal · 21g P · 21g C · 8g F
created food_log_entry: Water ×1 (240 g) 0 cal · 0g P · 0g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:51:42.921Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log miso soup and edamame." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said How much should I log for Miso soup and Edamame? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Miso soup and Edamame? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:51:54.615Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Miso soup" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Breakfast was granola, yogurt, and coffee." (difficulty —)0.5s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.5s

1 · TTS said Logged Granola, Plain Greek yogurt, and Coffee. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

2 · Card shown Logged Granola, Plain Greek yogurt, and Coffee. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

3 · App data rows written created food_log_entry: Granola ×1 (100 g) 471 cal · 10g P · 64g C · 20g F
created food_log_entry: Plain Greek yogurt ×1 (100 g) 97 cal · 9g P · 3.9g C · 5g F
created food_log_entry: Coffee ×1 (100 g) 1 cal · 0.1g P · 0g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:52:06.256Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"I ate a pear, string cheese, and pretzels." (difficulty —)0.2s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: Logged a BLIND guess — no stated assumption, no correction invited.

Why verdict Logged a BLIND guess — no stated assumption, no correction invited.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said How much should I log for Mozzarella string cheese and Pretzels? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Mozzarella string cheese and Pretzels? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:52:17.654Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"Dinner was pork, farro, and Brussels sprouts." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: Logged a BLIND guess — no stated assumption, no correction invited.

Why verdict Logged a BLIND guess — no stated assumption, no correction invited.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said How much should I log for Pork, Cooked farro, and Brussels sprouts? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Pork, Cooked farro, and Brussels sprouts? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:52:29.192Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Brussels sprouts" at 90% confidence.
Matched your past food logs to "Brussels sprouts" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Snack was snap peas and hummus." (difficulty —)0.3s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said How much should I log for Snap peas and Hummus? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Snap peas and Hummus? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:52:40.658Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Snap peas" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Lunch was a grain bowl and iced tea." (difficulty —)1.3s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.3s

1 · TTS said Logged Grain Bowl (serving) and Unsweetened iced tea. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

2 · Card shown Logged Grain Bowl (serving) and Unsweetened iced tea. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

3 · App data rows written created food_log_entry: Grain Bowl ×1 (serving) 313 cal · 8.1g P · 55.7g C · 6.7g F
created food_log_entry: Unsweetened iced tea ×1 (240 g) 2 cal · 0g P · 0g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:52:53.082Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had rice, lentils, and tofu." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said How much should I log for Cooked white rice, Lentils, and Firm tofu? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Cooked white rice, Lentils, and Firm tofu? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:53:04.722Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Cooked white rice" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Track skyr, honey, and walnuts." (difficulty —)0.6s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.6s

1 · TTS said How much should I log for Skyr, Honey, and Walnuts? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Skyr, Honey, and Walnuts? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:53:16.463Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Walnuts" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add eggs, Canadian bacon, and grapefruit juice." (difficulty —)0.6s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.6s

1 · TTS said How much should I log for Egg, Canadian bacon, and Grapefruit juice? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Egg, Canadian bacon, and Grapefruit juice? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:53:28.194Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Grapefruit juice" at 90% confidence.
Matched your past food logs to "Canadian bacon" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"I ate noodles and a side of spinach." (difficulty —)0.7s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: WRONG item vs seeded default/key — WRONG/MISSING item "noodles" — no saved row with matching identity (rows: Cooked pasta, Spinach, Cooked pasta, Spinach)

Why verdict WRONG item vs seeded default/key — WRONG/MISSING item "noodles" — no saved row with matching identity (rows: Cooked pasta, Spinach, Cooked pasta, Spinach)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.7s

1 · TTS said Logged Cooked pasta and Spinach. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

2 · Card shown Logged Cooked pasta and Spinach. Assumed catalog default servings where you did not say an amount — tell me if that is not right.

3 · App data rows written created food_log_entry: Cooked pasta ×1 (140 g) 220 cal · 8.1g P · 43.3g C · 1.3g F
created food_log_entry: Spinach ×1 (30 g) 7 cal · 0.9g P · 1.1g C · 0.1g F
created food_log_entry: Cooked pasta ×1 (140 g) 220 cal · 8.1g P · 43.3g C · 1.3g F
created food_log_entry: Spinach ×1 (30 g) 7 cal · 0.9g P · 1.1g C · 0.1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:53:40.102Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log a nectarine and almond butter." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said How much should I log for Almond butter? I did not log them yet because the amounts were not clear.

2 · Card shown How much should I log for Almond butter? I did not log them yet because the amounts were not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:53:51.827Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Almond butter" at 90% confidence.
Matched your past food logs to "Almond butter" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.6 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"Breakfast was cereal and oat milk." (difficulty —)1.2s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: Logged a BLIND guess — no stated assumption, no correction invited.

Why verdict Logged a BLIND guess — no stated assumption, no correction invited.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.2s

1 · TTS said I need to resolve cereal before I log this meal. What should I use for cereal?

2 · Card shown I need to resolve cereal before I log this meal. What should I use for cereal?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:54:04.293Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I had a frittata with four eggs, spinach, and feta" (difficulty —)2.5s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Egg, Frittata, Spinach, Feta cheese — card not captured.

Why verdict Logged (write-truth verified): Egg, Frittata, Spinach, Feta cheese — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.5s

1 · TTS said Logged Frittata. Includes four eggs, spinach, and Feta cheese.

2 · Card shown Logged Frittata. Includes four eggs, spinach, and Feta cheese.

3 · App data rows written created food_log_entry: Frittata ×1 (serving) 557 cal · 40.3g P · 6.6g C · 40.4g F
created food_log_entry: Egg ×1 (4 eggs) 286 cal · 25.2g P · 1.4g C · 19g F
created food_log_entry: Spinach ×1 (30 g) 7 cal · 0.9g P · 1.1g C · 0.1g F
created food_log_entry: Feta cheese ×1 (100 g) 264 cal · 14.2g P · 4.1g C · 21.3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:54:17.952Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"Log a smoothie with spinach, banana, and whey protein." (difficulty —)1.2s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: Logged a BLIND guess — no stated assumption, no correction invited.

Why verdict Logged a BLIND guess — no stated assumption, no correction invited.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.2s

1 · TTS said Logged Smoothie. Includes spinach, one banana, and Whey protein powder.

2 · Card shown Logged Smoothie. Includes spinach, one banana, and Whey protein powder.

3 · App data rows written created food_log_entry: Smoothie ×1 (serving) 222 cal · 27.1g P · 29.2g C · 1g F
created food_log_entry: Spinach ×1 (30 g) 7 cal · 0.9g P · 1.1g C · 0.1g F
created food_log_entry: Banana ×1 (1 banana) 105 cal · 1.3g P · 27.1g C · 0.4g F
created food_log_entry: Whey protein powder ×1 (30 g) 110 cal · 24.9g P · 1g C · 0.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:54:30.308Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"I had a grain bowl with quinoa, chickpeas, and tahini." (difficulty —)3.6s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: Logged a BLIND guess — no stated assumption, no correction invited.

Why verdict Logged a BLIND guess — no stated assumption, no correction invited.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.6s

1 · TTS said Logging your grain bowl with quinoa, chickpeas, and tahini — still looking up the nutrition details; they'll appear in your log shortly.

2 · Card shown Logging your grain bowl with quinoa, chickpeas, and tahini — still looking up the nutrition details; they'll appear in your log shortly.

3 · App data rows written pending_created food_log_entry: a grain bowl with quinoa, chickpeas, and tahini (pending)

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:54:52.056Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Breakfast was avocado toast with egg and chili flakes." (difficulty —)0.8s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.8s

1 · TTS said What exact ingredient should I use for chili flakes?

2 · Card shown What exact ingredient should I use for chili flakes?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:56:43.828Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Toast" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I made stir fry with tofu, peppers, broccoli, and oil." (difficulty —)1.0s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.0s

1 · TTS said What should I use for tofu?

2 · Card shown What should I use for tofu?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:56:55.891Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Dinner was a poke bowl with rice, tuna, avocado, and seaweed." (difficulty —)1.3s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.3s

1 · TTS said What should I use for tuna?

2 · Card shown What should I use for tuna?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:57:20.391Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"I had pho with beef and basil." (difficulty —)2.3s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Pho, Basil, Raw, Beef — card not captured.

Why verdict Logged (write-truth verified): Pho, Basil, Raw, Beef — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.3s

1 · TTS said Logged Pho. Includes Beef and Basil, Raw.

2 · Card shown Logged Pho. Includes Beef and Basil, Raw.

3 · App data rows written created food_log_entry: Pho ×1 (serving) 219 cal · 22.9g P · 0.6g C · 12.9g F
created food_log_entry: Beef ×1 (85 g) 213 cal · 22.1g P · 0g C · 12.8g F
created food_log_entry: Basil, Raw ×1 (1 cup) 6 cal · 0.8g P · 0.6g C · 0.1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:57:33.842Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould log the entry"Log a chicken wrap with hummus and cucumber." (difficulty —)2.5s

Verdict Expected LOG — should log the entry. PASS: Logged (write-truth verified): Cucumber, Chicken Wrap, Hummus, Plain — card not captured.

Why verdict Logged (write-truth verified): Cucumber, Chicken Wrap, Hummus, Plain — card not captured.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.5s

1 · TTS said Logged Chicken Wrap. Includes Hummus, Plain and Cucumber.

2 · Card shown Logged Chicken Wrap. Includes Hummus, Plain and Cucumber.

3 · App data rows written created food_log_entry: Chicken Wrap ×1 (serving) 51 cal · 1.8g P · 5.8g C · 2.7g F
created food_log_entry: Hummus, Plain ×1 (1 tablespoon) 36 cal · 1.1g P · 2.2g C · 2.6g F
created food_log_entry: Cucumber ×1 (100 g) 15 cal · 0.7g P · 3.6g C · 0.1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:57:47.533Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I built an acai bowl with granola, banana, and coconut." (difficulty —)3.9s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.9s

1 · TTS said Which coconut should I use?

2 · Card shown Which coconut should I use?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:58:02.589Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"Dinner was butter chicken with rice, naan, and cucumber salad." (difficulty —)0.9s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.9s

1 · TTS said What exact ingredient should I use for butter chicken with rice naan?

2 · Card shown What exact ingredient should I use for butter chicken with rice naan?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:58:14.600Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "cucumber" at 90% confidence.
Matched the common-food list to "Butter" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add some porridge." (difficulty —)1.0s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.0s

1 · TTS said How much some porridge did you have?

2 · Card shown How much some porridge did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:58:26.781Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had muesli." (difficulty —)0.8s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.8s

1 · TTS said Did you have milk, yogurt, fruit, or honey with the cereal or granola? I did not log it yet because add-ons can change the nutrition.

2 · Card shown Did you have milk, yogurt, fruit, or honey with the cereal or granola? I did not log it yet because add-ons can change the nutrition.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:58:39.438Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log couscous for dinner." (difficulty —)1.6s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.6s

1 · TTS said How much couscous did you have?

2 · Card shown How much couscous did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:58:52.215Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Track noodles." (difficulty —)2.3s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.3s

1 · TTS said How much noodles did you have?

2 · Card shown How much noodles did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:59:05.737Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I ate pretzels." (difficulty —)0.1s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said How much should I log for Pretzels? I did not log it yet because the amount was not clear.

2 · Card shown How much should I log for Pretzels? I did not log it yet because the amount was not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:59:16.941Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Pretzels" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add walnuts." (difficulty —)0.2s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said How much should I log for Walnuts? I did not log it yet because the amount was not clear.

2 · Card shown How much should I log for Walnuts? I did not log it yet because the amount was not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:59:28.416Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Walnuts" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log a bowl of chili." (difficulty —)2.7s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.7s

1 · TTS said What exact food and amount should I use for a bowl of chili? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for a bowl of chili? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:59:42.812Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had a plate of risotto." (difficulty —)2.7s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.7s

1 · TTS said What exact food and amount should I use for a plate of risotto? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for a plate of risotto? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T21:59:56.632Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Track lemonade." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said What exact food and amount should I use for lemonade? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for lemonade? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:00:08.304Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add tea." (difficulty —)0.5s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.5s

1 · TTS said How much tea did you have?

2 · Card shown How much tea did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:00:20.003Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I drank beer." (difficulty —)0.9s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.9s

1 · TTS said What exact food and amount should I use for beer? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for beer? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:00:32.039Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log gelato." (difficulty —)0.2s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said What food did you log? Could you be more specific?

2 · Card shown What food did you log? Could you be more specific?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:00:43.327Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Had some mixed nuts." (difficulty —)0.1s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said How much should I log for Mixed nuts? I did not log it yet because the amount was not clear.

2 · Card shown How much should I log for Mixed nuts? I did not log it yet because the amount was not clear.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:00:54.985Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Mixed nuts" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add kettle corn." (difficulty —)4.3s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 4.3s

1 · TTS said How much kettle corn did you have?

2 · Card shown How much kettle corn did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:01:11.435Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"I had an energy bar." (difficulty —)0.3s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: WRONG item vs seeded default/key — WRONG/MISSING item "energy bar" — no saved row with matching identity (rows: Generic protein bar, Generic protein bar, Generic protein bar)

Why verdict WRONG item vs seeded default/key — WRONG/MISSING item "energy bar" — no saved row with matching identity (rows: Generic protein bar, Generic protein bar, Generic protein bar)

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Logged an energy bar. Assumed 1 energy bar as the default serving. Tell me if you want a different amount.

2 · Card shown Logged an energy bar. Assumed 1 energy bar as the default serving. Tell me if you want a different amount.

3 · App data rows written created food_log_entry: Generic protein bar ×1 (1 energy bar) 200 cal · 20g P · 22g C · 7g F
created food_log_entry: Generic protein bar ×1 (1 energy bar) 200 cal · 20g P · 22g C · 7g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:01:22.752Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Generic protein bar" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log a dairy cup." (difficulty —)0.9s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.9s

1 · TTS said What food did you have?

2 · Card shown What food did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:01:46.428Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Track a milk alternative." (difficulty —)3.5s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.5s

1 · TTS said Which one did you mean? 1. Low Fat Milk Alternative — AUGASON FARMS (17 g), 2. Low Fat Milk Alternative — AUGASON FARMS (35 g)

2 · Card shown Which one did you mean? 1. Low Fat Milk Alternative — AUGASON FARMS (17 g), 2. Low Fat Milk Alternative — AUGASON FARMS (35 g)

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:02:01.140Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"I ate a deli sandwich." (difficulty —)1.4s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: Logged a BLIND guess — no stated assumption, no correction invited.

Why verdict Logged a BLIND guess — no stated assumption, no correction invited.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.4s

1 · TTS said Logged tuna melt sandwich with an estimated macro range.

2 · Card shown Logged tuna melt sandwich — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Tuna melt sandwich ×1 (1 tuna melt) 550 cal · 30.8g P · 44g C · 26.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:02:13.696Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add a protein shake." (difficulty —)3.9s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.9s

1 · TTS said I couldn't find Add A Protein Protein Shake in my nutrition database. What exact name is on the package? I did not log it because I'd only be guessing the nutrition. If you prefer, tell me to log a generic protein shake.

2 · Card shown I couldn't find Add A Protein Protein Shake in my nutrition database. What exact name is on the package? I did not log it because I'd only be guessing the nutrition. If you prefer, tell me to log a generic protein shake.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:02:28.764Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had a latte." (difficulty —)0.9s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.9s

1 · TTS said Got it, I logged latte.

2 · Card shown I logged latte — let me know if that's not right.

3 · App data rows written created food_log_entry: latte ×1 (serving) 150 cal · 10g P · 15g C · 5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:02:40.814Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log a roll." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said Got it, I logged Roll, White, Hot Dog Bun (1 roll (1 hot dog bun)).

2 · Card shown I logged Roll, White, Hot Dog Bun (1 roll (1 hot dog bun)) — let me know if that's not right.

3 · App data rows written created food_log_entry: Roll, White, Hot Dog Bun ×1 (1 roll (1 hot dog bun)) 120 cal · 4.2g P · 22.1g C · 1.6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:02:52.381Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Track a veggie burger." (difficulty —)3.0s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.0s

1 · TTS said What exact food and amount should I use for a veggie burger? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for a veggie burger? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:03:06.995Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had flatbread pizza." (difficulty —)4.5s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Asked a clarification as expected.

Why verdict Asked a clarification as expected.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 4.5s

1 · TTS said How much flatbread pizza did you have?

2 · Card shown How much flatbread pizza did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:03:22.185Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould ask a clarifying question"Add a scone." (difficulty —)0.4s

Verdict Expected CLARIFY — should ask a clarifying question. FAIL: Logged a BLIND guess — no stated assumption, no correction invited.

Why verdict Logged a BLIND guess — no stated assumption, no correction invited.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.4s

1 · TTS said Logged Scone, with Fruit.

2 · Card shown Logged Scone, with Fruit.

3 · App data rows written created food_log_entry: Scone, with Fruit ×1 (1 regular) 262 cal · 4.4g P · 35.8g C · 11.6g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:03:33.714Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Log my usual latte size." (difficulty —)0.1s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged latte. Using your saved "log my usual latte size". Tell me if that is wrong.

2 · Card shown Logged latte. Using your saved "log my usual latte size". Tell me if that is wrong.

3 · App data rows written created food_log_entry: latte ×1 (serving) 150 cal · 10g P · 15g C · 5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: latte

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-latte-confirm",
    "name": "latte",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 10,
      "fatGrams": 5,
      "carbsGrams": 15
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "log my usual latte size",
  "foodItemId": "seed-d-latte-confirm",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "log my usual latte size",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T22:03:45.047Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "log my usual latte size" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Add the regular dinner portion." (difficulty —)0.1s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged dinner portion. Using your saved "add the regular dinner portion". Tell me if that is wrong.

2 · Card shown Logged dinner portion. Using your saved "add the regular dinner portion". Tell me if that is wrong.

3 · App data rows written created food_log_entry: dinner portion ×1 (serving) 150 cal · 10g P · 15g C · 5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: dinner portion

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-dinner-portion",
    "name": "dinner portion",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 10,
      "fatGrams": 5,
      "carbsGrams": 15
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "add the regular dinner portion",
  "foodItemId": "seed-d-dinner-portion",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "add the regular dinner portion",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T22:03:56.442Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "add the regular dinner portion" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"Track my default afternoon snack." (difficulty —)0.1s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged snack. Using your saved "track my default afternoon snack". Tell me if that is wrong.

2 · Card shown Logged snack. Using your saved "track my default afternoon snack". Tell me if that is wrong.

3 · App data rows written created food_log_entry: snack ×1 (serving) 150 cal · 10g P · 15g C · 5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: snack

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-snack-amt",
    "name": "snack",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 10,
      "fatGrams": 5,
      "carbsGrams": 15
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "track my default afternoon snack",
  "foodItemId": "seed-d-snack-amt",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "track my default afternoon snack",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T22:04:07.753Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "track my default afternoon snack" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould ask a clarifying question"I had the usual breakfast size." (difficulty —)0.1s

Verdict Expected CLARIFY — should ask a clarifying question. PASS: Decisive guess with stated assumption + invited correction (good-guessing).

Why verdict Decisive guess with stated assumption + invited correction (good-guessing).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Logged breakfast. Using your saved "i had the usual breakfast size". Tell me if that is wrong.

2 · Card shown Logged breakfast. Using your saved "i had the usual breakfast size". Tell me if that is wrong.

3 · App data rows written created food_log_entry: breakfast ×1 (serving) 150 cal · 10g P · 15g C · 5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: breakfast

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-bfast-size",
    "name": "breakfast",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 10,
      "fatGrams": 5,
      "carbsGrams": 15
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "i had the usual breakfast size",
  "foodItemId": "seed-d-bfast-size",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "i had the usual breakfast size",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-08-02T22:04:19.020Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "i had the usual breakfast size" at 96% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"Three quarters of a cup." (difficulty —)2.4s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. UNVERIFIED: CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.4s

1 · TTS said What food did you have?

2 · Card shown What food did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:04:32.644Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"One and a half cups, plain." (difficulty —)1.4s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.4s

1 · TTS said I couldn't find A Half Half Packaged Food in my nutrition database. What exact name is on the package? I did not log it because I'd only be guessing the nutrition. If you prefer, tell me to log a generic packaged food.

2 · Card shown I couldn't find A Half Half Packaged Food in my nutrition database. What exact name is on the package? I did not log it because I'd only be guessing the nutrition. If you prefer, tell me to log a generic packaged food.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:04:45.218Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"The RXBAR blueberry." (difficulty —)1.0s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 1.0s

1 · TTS said How much The RXBAR blueberry did you have?

2 · Card shown How much The RXBAR blueberry did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:04:57.352Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"Oat milk." (difficulty —)4.1s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 4.1s

1 · TTS said How much Oat milk did you have?

2 · Card shown How much Oat milk did you have?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:05:12.671Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"No, that's not it — the unsweetened one." (difficulty —)0.1s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said What should I change about breakfast?

2 · Card shown What should I change about breakfast?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:05:23.892Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould complete the pending action"Wrong one. The vanilla, not chocolate." (difficulty —)2.7s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. FAIL: Did not complete the pending action.

Why verdict Did not complete the pending action.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.7s

1 · TTS said I didn't log any latte. Please use the exact app command to log your drink, like "Log a vanilla latte."

2 · Card shown I didn't log any latte. Please use the exact app command to log your drink, like "Log a vanilla latte."

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:05:37.763Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"No, not the flavored skyr." (difficulty —)0.1s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said What should I change about breakfast?

2 · Card shown What should I change about breakfast?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:05:49.034Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Skyr" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould decline"Cancel that entry." (difficulty —)0.1s

Verdict Expected DECLINE — should decline. PASS: Declined / no unwanted write.

Why verdict Declined / no unwanted write.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Should I add that as a to-do?

2 · Card shown Should I add that as a to-do?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Pending Food Item

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-corr-pending-3",
    "name": "Pending Food Item",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 10,
      "fatGrams": 5,
      "carbsGrams": 15
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-corr-pending-3 at 2026-07-18T11:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-corr-pending-3",
  "quantity": 1,
  "loggedAt": "2026-07-18T11:00:00.000Z"
}

snapshot captured 2026-08-02T22:06:11.844Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould decline"Never mind, remove that." (difficulty —)2.6s

Verdict Expected DECLINE — should decline. PASS: Declined / no unwanted write.

Why verdict Declined / no unwanted write.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.6s

1 · TTS said Understood. Your last entry has been cleared.

2 · Card shown Understood. Your last entry has been cleared.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Pending Food Item

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-corr-pending-4",
    "name": "Pending Food Item",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 10,
      "fatGrams": 5,
      "carbsGrams": 15
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-corr-pending-4 at 2026-07-18T11:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-corr-pending-4",
  "quantity": 1,
  "loggedAt": "2026-07-18T11:00:00.000Z"
}

snapshot captured 2026-08-02T22:06:25.560Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"No, that's not the bar I meant." (difficulty —)0.1s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said What should I change about breakfast?

2 · Card shown What should I change about breakfast?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:06:36.783Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"Not the medium — the large pear." (difficulty —)5.3s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 5.3s

1 · TTS said What exact food and amount should I use for Pear? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for Pear? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:06:53.281Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"Wrong item. Pick the baked cod." (difficulty —)3.7s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.7s

1 · TTS said What exact food and amount should I use for Cod? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for Cod? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:07:08.159Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"That's not it; the plain one." (difficulty —)2.8s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.8s

1 · TTS said What would you like me to do with that?

2 · Card shown What would you like me to do with that?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:07:33.321Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould complete the pending action"No, the other cheese." (difficulty —)6.5s

Why verdict CLARIFY_ANSWER not gradeable single-shot — no pending clarification in this replay (app re-asked / mis-parsed as fresh); needs multi-turn replay.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 6.5s

1 · TTS said Which exact other cheese should replace breakfast?

2 · Card shown Which exact other cheese should replace breakfast?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:07:50.964Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould complete the pending action"Not those cashews — the raw ones." (difficulty —)8.9s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. FAIL: Did not complete the pending action.

Why verdict Did not complete the pending action.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 8.9s

1 · TTS said Got it — raw cashews, not the others. To log them, say exactly: "Log raw cashews" followed by the amount.

2 · Card shown Got it — raw cashews, not the others. To log them, say exactly: "Log raw cashews" followed by the amount.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:08:22.197Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "cashews" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Delete the banana from earlier." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not find a list matching earlier. Which list has banana?

2 · Card shown I could not find a list matching earlier. Which list has banana?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:08:44.594Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Banana" at 90% confidence.
Matched your saved foods and aliases to "one banana" at 64% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Remove that Quest bar." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Which food entry did you want to correct?

2 · Card shown Which food entry did you want to correct?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:08:55.800Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "QUEST Cookies & Cream Protein Bar" at 77% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Delete the oats entry." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Which food entry did you want to correct?

2 · Card shown Which food entry did you want to correct?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:09:07.217Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Dry oats" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Take off the eggs." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Which food entry did you want to correct?

2 · Card shown Which food entry did you want to correct?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:09:18.464Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Egg" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Remove the raspberries." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Which food entry did you want to correct?

2 · Card shown Which food entry did you want to correct?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:09:29.661Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Raspberries" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Delete the chicken log." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Which food entry did you want to correct?

2 · Card shown Which food entry did you want to correct?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:09:40.886Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Chicken breast" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Scratch the almond butter." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Which food entry did you want to correct?

2 · Card shown Which food entry did you want to correct?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:09:52.132Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Almond butter" at 90% confidence.
Matched your past food logs to "Almond butter" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Remove the milk entry." (difficulty —)0.1s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Which food entry did you want to correct?

2 · Card shown Which food entry did you want to correct?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:10:03.349Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Milk" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould delete the entry"Delete that brownie I logged last night." (difficulty —)0.2s

Verdict Expected DELETE — should delete the entry. PASS: Deleted/removed the requested food with database removal proof.

Why verdict Deleted/removed the requested food with database removal proof.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Deleted brownie.

2 · Card shown Deleted brownie.

3 · App data rows written deleted food_log_entry:

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: brownie

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-del-0-brownie",
    "name": "brownie",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 200,
      "proteinGrams": 5,
      "fatGrams": 10,
      "carbsGrams": 20
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-del-0-brownie at 2026-07-17T20:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-del-0-brownie",
  "quantity": 1,
  "loggedAt": "2026-07-17T20:00:00.000Z"
}

snapshot captured 2026-08-02T22:10:21.364Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould delete the entry"Remove yesterday's gelato." (difficulty —)0.2s

Verdict Expected DELETE — should delete the entry. PASS: Deleted/removed the requested food with database removal proof.

Why verdict Deleted/removed the requested food with database removal proof.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Deleted gelato.

2 · Card shown Deleted gelato.

3 · App data rows written deleted food_log_entry:

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: gelato

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-del-1-gelato",
    "name": "gelato",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 200,
      "proteinGrams": 5,
      "fatGrams": 10,
      "carbsGrams": 20
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-del-1-gelato at 2026-07-17T20:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-del-1-gelato",
  "quantity": 1,
  "loggedAt": "2026-07-17T20:00:00.000Z"
}

snapshot captured 2026-08-02T22:10:44.119Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould delete the entry"Delete the leftover burrito entry." (difficulty —)0.2s

Verdict Expected DELETE — should delete the entry. PASS: Deleted/removed the requested food with database removal proof.

Why verdict Deleted/removed the requested food with database removal proof.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Deleted burrito.

2 · Card shown Deleted burrito.

3 · App data rows written deleted food_log_entry:

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: burrito

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-del-2-burrito",
    "name": "burrito",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 200,
      "proteinGrams": 5,
      "fatGrams": 10,
      "carbsGrams": 20
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-del-2-burrito at 2026-07-17T20:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-del-2-burrito",
  "quantity": 1,
  "loggedAt": "2026-07-17T20:00:00.000Z"
}

snapshot captured 2026-08-02T22:10:55.968Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould delete the entry"Take out the wine from last night." (difficulty —)3.4s

Verdict Expected DELETE — should delete the entry. FAIL: Did not delete (not understood / no removal).

Why verdict Did not delete (not understood / no removal).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.4s

1 · TTS said I did not make any app changes. What would you like me to do with that?

2 · Card shown I did not make any app changes. What would you like me to do with that?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: wine

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-del-3-wine",
    "name": "wine",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 200,
      "proteinGrams": 5,
      "fatGrams": 10,
      "carbsGrams": 20
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-del-3-wine at 2026-07-17T20:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-del-3-wine",
  "quantity": 1,
  "loggedAt": "2026-07-17T20:00:00.000Z"
}

snapshot captured 2026-08-02T22:11:03.514Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.7 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould delete the entry"Remove the pastry log from earlier this week." (difficulty —)0.2s

Verdict Expected DELETE — should delete the entry. PASS: Deleted/removed the requested food with database removal proof.

Why verdict Deleted/removed the requested food with database removal proof.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Deleted pastry.

2 · Card shown Deleted pastry.

3 · App data rows written deleted food_log_entry:

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: pastry

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-del-4-pastry",
    "name": "pastry",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 200,
      "proteinGrams": 5,
      "fatGrams": 10,
      "carbsGrams": 20
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-del-4-pastry at 2026-07-17T20:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-del-4-pastry",
  "quantity": 1,
  "loggedAt": "2026-07-17T20:00:00.000Z"
}

snapshot captured 2026-08-02T22:11:21.451Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould delete the entry"Delete the chips I shouldn't have logged." (difficulty —)0.2s

Verdict Expected DELETE — should delete the entry. PASS: Deleted/removed the requested food with database removal proof.

Why verdict Deleted/removed the requested food with database removal proof.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.2s

1 · TTS said Deleted chips.

2 · Card shown Deleted chips.

3 · App data rows written deleted food_log_entry:

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: chips

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-del-5-chips",
    "name": "chips",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 200,
      "proteinGrams": 5,
      "fatGrams": 10,
      "carbsGrams": 20
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-del-5-chips at 2026-07-17T20:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-del-5-chips",
  "quantity": 1,
  "loggedAt": "2026-07-17T20:00:00.000Z"
}

snapshot captured 2026-08-02T22:11:32.811Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Potato chips" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Change the sequence banana quantity to two." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:11:37.487Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Banana" at 90% confidence.
Matched your saved foods and aliases to "one banana" at 61% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Actually make the Quest bar two bars." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:11:48.707Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your past food logs to "QUEST Cookies & Cream Protein Bar" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Update the oats to one and a half cups." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:12:00.229Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Dry oats" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Change the eggs to four." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:12:11.244Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Egg" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Make the raspberries a half cup." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:12:22.446Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Raspberries" at 90% confidence.
Matched your past food logs to "Raspberries" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould update the entry"Update the chicken to seven ounces." (difficulty —)0.3s

Verdict Expected UPDATE — should update the entry. PASS: Updated as asked.

Why verdict Updated as asked.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Updated Chicken Wrap to 7 oz.

2 · Card shown Updated Chicken Wrap to 7 oz.

3 · App data rows written updated food_log_entry: Chicken Wrap ×13.229766666666666 (1 tablespoon) 476 cal · 14.6g P · 29.1g C · 34.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:12:33.977Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Chicken breast" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Change almond butter to one tablespoon." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:12:45.232Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Almond butter" at 90% confidence.
Matched your past food logs to "Almond butter" at 82% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould update the entry"Actually the milk was two percent, not whole." (difficulty —)3.1s

Verdict Expected UPDATE — should update the entry. PASS: Updated as asked.

Why verdict Updated as asked.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 3.1s

1 · TTS said Updated to two percent milk.

2 · Card shown Updated the food entry from "Oat Milk" to "two percent milk".

3 · App data rows written updated food_log_entry: Milk ×2 (100 g) 100 cal · 6.6g P · 9.6g C · 4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:12:59.517Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould update the entry"Actually that was strawberry, not vanilla." (difficulty —)0.3s

Verdict Expected UPDATE — should update the entry. PASS: Updated as asked.

Why verdict Updated as asked.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.3s

1 · TTS said Updated to strawberry.

2 · Card shown Updated the food entry from "breakfast" to "strawberry".

3 · App data rows written updated food_log_entry: Chobani Greek Yogurt, Strawberry, Zero Sugar ×1 (one) 60 cal · 12g P · 5g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: vanilla item

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-edit-0-vanilla-item",
    "name": "vanilla item",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 8,
      "fatGrams": 6,
      "carbsGrams": 18
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-edit-0-vanilla-item at 2026-07-18T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-edit-0-vanilla-item",
  "quantity": 1,
  "loggedAt": "2026-07-18T10:00:00.000Z"
}

snapshot captured 2026-08-02T22:13:10.992Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Strawberries" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.5 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Change the portion to one and a half cups." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: quinoa

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-edit-1-quinoa",
    "name": "quinoa",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 8,
      "fatGrams": 6,
      "carbsGrams": 18
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-edit-1-quinoa at 2026-07-18T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-edit-1-quinoa",
  "quantity": 1,
  "loggedAt": "2026-07-18T10:00:00.000Z"
}

snapshot captured 2026-08-02T22:13:22.215Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Make it a quarter cup instead." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: granola

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-edit-2-granola",
    "name": "granola",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 8,
      "fatGrams": 6,
      "carbsGrams": 18
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-edit-2-granola at 2026-07-18T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-edit-2-granola",
  "quantity": 1,
  "loggedAt": "2026-07-18T10:00:00.000Z"
}

snapshot captured 2026-08-02T22:13:33.501Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould update the entry"Update that to eight ounces." (difficulty —)0.1s

Verdict Expected UPDATE — should update the entry. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I could not confirm that was saved, so I did not mark it as done.

2 · Card shown I could not confirm that the change was saved, so I did not mark it as done.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: salmon

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-edit-3-salmon",
    "name": "salmon",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 8,
      "fatGrams": 6,
      "carbsGrams": 18
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-edit-3-salmon at 2026-07-18T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-edit-3-salmon",
  "quantity": 1,
  "loggedAt": "2026-07-18T10:00:00.000Z"
}

snapshot captured 2026-08-02T22:13:44.671Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould update the entry"Actually it was cashew butter, not almond butter." (difficulty —)2.5s

Verdict Expected UPDATE — should update the entry. PASS: Updated as asked.

Why verdict Updated as asked.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 2.5s

1 · TTS said Updated to cashew butter.

2 · Card shown Updated the food entry from "Chobani Greek Yogurt, Strawberry, Zero Sugar" to "cashew butter".

3 · App data rows written updated food_log_entry: Cashew Butter ×1 (1 tablespoon) 97 cal · 1.9g P · 4.8g C · 8.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: almond butter

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-edit-4-almond-butter",
    "name": "almond butter",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 8,
      "fatGrams": 6,
      "carbsGrams": 18
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-edit-4-almond-butter at 2026-07-18T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-edit-4-almond-butter",
  "quantity": 1,
  "loggedAt": "2026-07-18T10:00:00.000Z"
}

snapshot captured 2026-08-02T22:13:58.461Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould update the entry"Change the pear to a small." (difficulty —)5.8s

Verdict Expected UPDATE — should update the entry. PASS: Updated as asked.

Why verdict Updated as asked.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 5.8s

1 · TTS said Updated Pear to 1 small.

2 · Card shown Updated Pear to 1 small.

3 · App data rows written updated food_log_entry: Pear ×1 (178 g) 101 cal · 0.7g P · 27.1g C · 0.2g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: pear

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-d-edit-5-pear",
    "name": "pear",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 8,
      "fatGrams": 6,
      "carbsGrams": 18
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-d-edit-5-pear at 2026-07-18T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-d-edit-5-pear",
  "quantity": 1,
  "loggedAt": "2026-07-18T10:00:00.000Z"
}

snapshot captured 2026-08-02T22:14:21.989Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"How much protein is left for today?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said You are about 898.1 grams of protein over today's 160 gram protein target (1058.1 logged).

2 · Card shown You are about 898.1 grams of protein over today's 160 gram protein target (1058.1 logged).

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:14:26.851Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"How many calories have I logged?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I do not see calories in today's food log.

2 · Card shown I do not see calories in today's food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:14:49.864Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"What did I eat last night?" (difficulty —)0.0s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.0s

1 · TTS said I do not see any food logged yesterday.

2 · Card shown I do not see any food logged yesterday.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:15:01.284Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"How much fiber today?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

2 · Card shown I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:15:12.212Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"What's left in my calorie budget?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said You are about 16492 calories over today's 2200 calorie budget (18692 logged).

2 · Card shown You are about 16492 calories over today's 2200 calorie budget (18692 logged).

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:15:23.460Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"Did I log lunch yet?" (difficulty —)0.0s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.0s

1 · TTS said I do not see lunch logged today.

2 · Card shown I do not see lunch logged today.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:15:34.614Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"How many carbs so far?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

2 · Card shown I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:15:45.798Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"What did I eat before the gym?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said No — I do not see before the gym logged today.

2 · Card shown No — I do not see before the gym logged today.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:15:56.994Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.1 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"Show me yesterday's food log." (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Opening Food History.

2 · Card shown Opening Food History.

4 · UI did navigated to "food" screen

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:16:08.285Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"Am I under my fat goal?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

2 · Card shown I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:16:19.516Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"How many calories in the pear I just logged?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Pear has about 101 calories per 178 g.

2 · Card shown Pear has about 101 calories per 178 g.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:16:30.923Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched the common-food list to "Pear" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"What was my first meal today?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Your first food entry today was 1 1 large banana Banana.

2 · Card shown Your first food entry today was 1 1 large banana Banana.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:16:42.424Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"How much coffee have I logged?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Today you logged 1 100 g Coffee (1 calories).

2 · Card shown Today you logged 1 100 g Coffee (1 calories).

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:16:53.817Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "coffee" at 90% confidence.
Matched the common-food list to "Coffee" at 90% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.8 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"Calories left for a snack?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said You are about 16492 calories over today's 2200 calorie budget (18692 logged).

2 · Card shown You are about 16492 calories over today's 2200 calorie budget (18692 logged).

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:17:05.047Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 1.9 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"Did I already log my usual RXBAR?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Yes. Today you logged 1 bar RXBAR RXBAR Blueberry.

2 · Card shown Yes. Today you logged 1 bar RXBAR RXBAR Blueberry.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:17:16.466Z

6 · Why this food

The lookup compared against planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
Matched your saved foods and aliases to "my usual rxbar" at 96% confidence.
Matched your saved foods and aliases to "RXBAR Blueberry" at 80% confidence.

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 0.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"What's my average calories this week?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

2 · Card shown I understood the nutrition question, but I could not find a solid number for that food yet. Nothing was logged.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:17:27.751Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.3 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"How many food entries today?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said You have 89 food entries logged today.

2 · Card shown You have 89 food entries logged today.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:17:39.035Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

UNVERIFIEDunclassifiedshould answer the query"What did I have mid-afternoon?" (difficulty —)0.1s

Verdict Expected QUERY — should answer the query. UNVERIFIED: answer key has no expected values

Why verdict answer key has no expected values

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I do not see any food logged mid-afternoon today.

2 · Card shown I do not see any food logged mid-afternoon today.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:17:50.322Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

PASSshould navigate"Show the food diary screen." (difficulty —)0.1s

Verdict Expected NAVIGATE — should navigate. PASS: Navigated.

Why verdict Navigated.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said Opening Food.

2 · Card shown Opening Food.

4 · UI did navigated to "food" screen

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:18:01.553Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.2 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould navigate"Open macro goals." (difficulty —)0.1s

Verdict Expected NAVIGATE — should navigate. FAIL: Did not navigate — misrouted into a food/workout clarification or dead-ended.

Why verdict Did not navigate — misrouted into a food/workout clarification or dead-ended.

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said What would you like me to do with that?

2 · Card shown What would you like me to do with that?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:18:12.915Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.0 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould apply the setting"Set my protein goal to one hundred sixty grams." (difficulty —)0.1s

Verdict Expected SETTINGS — should apply the setting. FAIL: Reported could-not-confirm (write-confirmation false negative).

Why verdict Reported could-not-confirm (write-confirmation false negative).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 0.1s

1 · TTS said I'm not sure how to help with that yet. Try asking me to log food, a workout, a note, or a to-do.

2 · Card shown I'm not sure how to help with that yet. Try asking me to log food, a workout, a note, or a to-do.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:18:24.277Z

7 · Lookup method

The report captured the app food lookup that ran before or alongside the final decision.
It searched planned foods, your saved foods and aliases, your past food logs, the shared nutrition database, the common-food list, the exercise list, and the workout equipment list.
The lookup evidence took 2.4 seconds to collect.
This lookup is supporting evidence in the report; by itself it does not prove that food was saved.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

FAILshould log the entry"I drank a glass of orange juice." (difficulty —)5.4s

Verdict Expected LOG — should log the entry. FAIL: OVER-ASK — asked instead of logging (no saved row).

Why verdict OVER-ASK — asked instead of logging (no saved row).

My override

Things I think are wrong

Use this when the report passed something that should have failed.

Comprehension

Picked wrong food Wrong quantity Should have asked (over-logged) Shouldn't have asked (over-asked) Wrong action entirely

Execution

Didn't save to DB Wrong data saved No voice/TTS Card missing/wrong No chat

Cosmetic

Wording weird

Other thing wrong (describe below)

Things the report marked wrong but I think are right

Use this when the report failed something that should have passed.

Actually right

Food/action was right Quantity/serving was right Asking was right Not asking was right Answer was right No app data row was expected App data row was right Voice/card/chat was acceptable

Report/grader issue

Report expected the wrong result Evidence was enough to judge Failure reason was not valid

Other thing actually right (describe below)

Pick every reason you think applies. Multiple boxes are allowed because one utterance can be wrong in more than one way. Use the note below for "Other" or extra detail.

Timing 5.4s

1 · TTS said What exact food and amount should I use for Orange juice? I did not log it yet because I could not safely finish that food log.

2 · Card shown What exact food and amount should I use for Orange juice? I did not log it yet because I could not safely finish that food log.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state: none listed in the corpus row.

No DB snapshot captured for this utterance; the report cannot prove whether yesterday/personal data existed.

snapshot captured 2026-08-02T22:18:52.480Z

6 · Why this food No food-decision explanation was captured for this path.

7 · Lookup method No lookup method was captured for this path.

8 · Candidates → decider no candidate list captured (this path did not run DB resolution — e.g. provisional branded log)

Why the fails happened — comprehension vs execution vs cosmetic

Handled correctly? — by expected action

Accuracy by difficulty

Clarification follow-ups — scored separately

Cosmetic only

System / infra

Latency

Filter — controls the list below

Per-utterance detail