food-corpus-mixed-F-7c039dd0-c16-2…

Supposed to	N	Correct	Wrong	Unverified
▸ LOG — log the food	104	60 (58%)	41 (39%)	3 (3%)
41 handled wrong — click one to jump to its full detail below "Log one large tangerine for the sequence." OVER-ASK — asked instead of logging (no saved row). "Track one Aloha chocolate chip cookie dough bar." OVER-ASK — asked instead of logging (no saved row). "I had one cup grits with water only." OVER-ASK — asked instead of logging (no saved row). "Add two sunny side eggs for lunch." OVER-ASK — asked instead of logging (no saved row). "Track one tablespoon cashew butter." WRITE-TRUTH FAIL — MACROS off for "cashew butter": logged 180 kcal vs expected ~94 (§14 15pct/25cal band) "Breakfast was two sunny side eggs." OVER-ASK — asked instead of logging (no saved row). "Snack was one large tangerine." Reported could-not-confirm (write-confirmation false negative). "Count eighteen pistachios." OVER-ASK — asked instead of logging (no saved row). "Lunch included four ounces roast pork." OVER-ASK — asked instead of logging (no saved row). "I drank eight ounces pineapple juice." WRITE-TRUTH FAIL — MACROS off for "pineapple juice": logged 720 kcal vs expected ~132 (§14 15pct/25cal band) "Note two chicken breakfast sausages." WRITE-TRUTH FAIL — expected food write missing — local foodEntries delta is 0 for this turn (proven no food write, not a capture gap) "Supper was one cup kidney beans." OVER-ASK — asked instead of logging (no saved row). "Enter one medium apricot." Reported could-not-confirm (write-confirmation false negative). "Track one tablespoon cashew butter." WRITE-TRUTH FAIL — MACROS off for "cashew butter": logged 180 kcal vs expected ~94 (§14 15pct/25cal band) "Put in one hundred eighty grams skyr-style yogurt." OVER-ASK — asked instead of logging (no saved row). "Add one cup black-eyed peas." OVER-ASK — asked instead of logging (no saved row). "Log two frozen waffles whole grain." OVER-ASK — asked instead of logging (no saved row). "I ate one plain croissant small." OVER-ASK — asked instead of logging (no saved row). "Log one medium sweet onion sauteed." OVER-ASK — asked instead of logging (no saved row). "Track eight ounces cold brew coffee." OVER-ASK — asked instead of logging (no saved row). "Dinner was six ounces turkey meatballs." WRITE-TRUTH FAIL — MACROS off for "turkey meatballs": logged 315 kcal vs expected ~220 (§14 15pct/25cal band) "Put down one cup bulgur." OVER-ASK — asked instead of logging (no saved row). "Add my usual Aloha bar." OVER-ASK — asked instead of logging (no saved row). "Add the same ahi I had earlier." OVER-ASK — asked instead of logging (no saved row). "Log the shrimp I had last time." OVER-ASK — asked instead of logging (no saved row). "Snack was one Larabar cashew cookie." Reported could-not-confirm (write-confirmation false negative). "Finished one Fairlife Core Power banana shake." WRITE-TRUTH FAIL — expected food write missing — local foodEntries delta is 0 for this turn (proven no food write, not a capture gap) "Track one Fage total zero strawberry cup." OVER-ASK — asked instead of logging (no saved row). "Lunch was a turkey burger and a tangerine." Reported could-not-confirm (write-confirmation false negative). "Breakfast was oatmeal, yogurt, and coffee." Reported could-not-confirm (write-confirmation false negative). "Breakfast was cereal and almond milk." Reported could-not-confirm (write-confirmation false negative). "I had huevos rancheros with two eggs, salsa, and black beans" OVER-ASK — asked instead of logging (no saved row). "Log a smoothie with cherries, kale, and casein protein." WRITE-TRUTH FAIL — COMPOUND: component name missing — expected like "casein", got: Cherries, Protein Bar, Kale, Raw "I had a grain bowl with bulgur, lentils, and yogurt sauce." OVER-ASK — asked instead of logging (no saved row). "Breakfast was breakfast burrito with egg, potato, and cheese." OVER-ASK — asked instead of logging (no saved row). "I made jambalaya with shrimp, sausage, rice, and peppers." OVER-ASK — asked instead of logging (no saved row). "Dinner was a taco bowl with rice, carnitas, pico, and guacamole." Reported could-not-confirm (write-confirmation false negative). "I had ramen with chicken and corn." OVER-ASK — asked instead of logging (no saved row). "Log a chicken sandwich with pickles and mustard." OVER-ASK — asked instead of logging (no saved row). "I built an acai bowl with granola, pineapple, and coconut." OVER-ASK — asked instead of logging (no saved row). "Dinner was green curry with tofu, eggplant, and jasmine rice." OVER-ASK — asked instead of logging (no saved row).
▸ CLARIFY — ask a clarifying question	29	27 (93%)	2 (7%)	0 (0%)
2 handled wrong — click one to jump to its full detail below "I ate tortilla chips." Logged a BLIND guess — no stated assumption, no correction invited. "I had an Aloha bar." Logged a BLIND guess — no stated assumption, no correction invited.
▸ CLARIFY_ANSWER — complete the pending action	13	3 (23%)	6 (46%)	4 (31%)
6 handled wrong — click one to jump to its full detail below "No, that's not it — the unsweetened one." Reported could-not-confirm (write-confirmation false negative). "No, not the flavored labneh." Reported could-not-confirm (write-confirmation false negative). "No, that's not the bar I meant." Reported could-not-confirm (write-confirmation false negative). "Not the large — the small tangerine." Reported could-not-confirm (write-confirmation false negative). "That's not it; the plain one." Reported could-not-confirm (write-confirmation false negative). "No, the other tortilla." Reported could-not-confirm (write-confirmation false negative).
▸ DECLINE — decline	2	2 (100%)	0 (0%)	0 (0%)
No errors — all handled correctly.
Total	148	92 (65%)	49	7

FAILshould log the food"Log one large tangerine for the sequence." (difficulty —)2.7s

Verdict Expected LOG — should log the food. 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 2.7s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for one large tangerine for the sequence? I did not log it yet because I could not safely finish that food log.

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 2026-07-30T20:23:24.235Z

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 food"Track one Aloha chocolate chip cookie dough bar." (difficulty —)0.5s

Verdict Expected LOG — should log the food. 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 no speech captured

2 · Card shown Which one did you mean? 1. Chocolate Chip Cookie Dough Protein Bar — Aloha, 2. Chocolate Chip Cookie Dough Organic Protein Bar — ALOHA

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 2026-07-30T20:23:35.884Z

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 food"I had one cup grits with water only." (difficulty —)2.8s

Verdict Expected LOG — should log the food. 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 2.8s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for Instant grits? 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 2026-07-30T20:23:49.833Z

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 food"Add two sunny side eggs for lunch." (difficulty —)5.7s

Verdict Expected LOG — should log the food. 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 5.7s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for Egg? 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 2026-07-30T20:24:06.683Z

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 food"Log one cup cherries as a snack." (difficulty —)0.3s

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

Why verdict Logged (write-truth verified): Cherries — 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 no speech captured

2 · Card shown Logged one cup cherries 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: Cherries ×1 (1 cup) 97 cal · 1.7g P · 24.6g 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 2026-07-30T20:24:18.118Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Record six ounces grilled shrimp." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Shrimp — 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 no speech captured

2 · Card shown Logged six ounces grilled shrimp.

3 · App data rows written created food_log_entry: Shrimp ×1 (6 oz) 168 cal · 40.8g P · 0.3g 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 2026-07-30T20:24:29.417Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Track one tablespoon cashew butter." (difficulty —)0.2s

Verdict Expected LOG — should log the food. FAIL: WRITE-TRUTH FAIL — MACROS off for "cashew butter": logged 180 kcal vs expected ~94 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "cashew butter": logged 180 kcal vs expected ~94 (§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 no speech captured

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

3 · App data rows written created food_log_entry: cashew butter ×1 (serving) 180 cal · 8g P · 18g 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 2026-07-30T20:24:40.716Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"I drank one cup unsweetened almond milk." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Unsweetened almond 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 no speech captured

2 · Card shown Logged one cup unsweetened almond 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: Unsweetened almond milk ×1 (1 cup) 31 cal · 1g P · 0.7g C · 2.6g 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 2026-07-30T20:24:52.034Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I had one medium papaya half." (difficulty —)0.3s

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

Why verdict Logged (write-truth verified): Papaya — 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 no speech captured

2 · Card shown Logged one medium papaya half. Assumed 100 g because no amount was stated.

3 · App data rows written created food_log_entry: Papaya ×1 (100 g) 43 cal · 0.5g P · 10.8g 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 2026-07-30T20:25:03.434Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Breakfast was two sunny side eggs." (difficulty —)5.4s

Verdict Expected LOG — should log the food. 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 no speech captured

2 · Card shown What exact food and amount should I use for Egg? 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 2026-07-30T20:25:19.933Z

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 food"Put down six ounces grilled shrimp." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Shrimp — 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 no speech captured

2 · Card shown Logged six ounces grilled shrimp.

3 · App data rows written created food_log_entry: Shrimp ×1 (6 oz) 168 cal · 40.8g P · 0.3g 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 2026-07-30T20:25:31.200Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

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

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

Why verdict Logged (write-truth verified): Cooked jasmine 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 no speech captured

2 · Card shown Logged one cup cooked jasmine 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 jasmine rice ×1 (1 cup) 205 cal · 3.8g P · 44.6g 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 2026-07-30T20:25:42.584Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Snack was one large tangerine." (difficulty —)1.7s

Verdict Expected LOG — should log the food. 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 1.7s

1 · TTS said no speech captured

2 · Card shown I could not finish that request safely.

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 2026-07-30T20:25:55.517Z

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 log the food"I ate five ounces seared ahi tuna." (difficulty —)0.1s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Seared Ahi Tuna — card not captured.

Why verdict Logged (write-truth verified): Seared Ahi Tuna — 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 no speech captured

2 · Card shown Logged Seared Ahi Tuna (five ounces (141.7 g)).

3 · App data rows written created food_log_entry: Seared Ahi Tuna ×1 (five ounces (141.7 g)) 184 cal · 41.1g P · 0g 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 2026-07-30T20:26:06.800Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Track one cup cherries." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Cherries — 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 no speech captured

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

3 · App data rows written created food_log_entry: Cherries ×1 (1 cup) 97 cal · 1.7g P · 24.6g 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 2026-07-30T20:26:18.084Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Count eighteen pistachios." (difficulty —)0.1s

Verdict Expected LOG — should log the food. 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 0.1s

1 · TTS said no speech captured

2 · Card shown How much should I log for Pistachios? 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 2026-07-30T20:26:29.367Z

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 food"Lunch included four ounces roast pork." (difficulty —)5.3s

Verdict Expected LOG — should log the food. 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 5.3s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for Pork? 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 2026-07-30T20:26:45.784Z

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 food"Save one cup roasted carrots." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Baby carrots — 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 no speech captured

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

3 · App data rows written created food_log_entry: Baby carrots ×1 (1 cup) 52 cal · 0.9g P · 12.2g 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 2026-07-30T20:26:57.083Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"I drank eight ounces pineapple juice." (difficulty —)1.6s

Verdict Expected LOG — should log the food. FAIL: WRITE-TRUTH FAIL — MACROS off for "pineapple juice": logged 720 kcal vs expected ~132 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "pineapple juice": logged 720 kcal vs expected ~132 (§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 1.6s

1 · TTS said no speech captured

2 · Card shown I logged Klass, Flavored Drink Mix, With Pineapple Juice, Pina Colada — Grupo Dilcomer, S.A. De C.V. (eight ounces (176 g)) — let me know if that's not right.

3 · App data rows written created food_log_entry: Klass, Flavored Drink Mix, With Pineapple Juice, Pina Colada ×1 (eight ounces (176 g)) 720 cal · 0g P · 176g 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 2026-07-30T20:27:09.949Z

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

7 · Lookup method The answer used provisional food evidence from the USDA nutrition database.

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

FAILshould log the food"Note two chicken breakfast sausages." (difficulty —)0.3s

Verdict Expected LOG — should log the food. FAIL: WRITE-TRUTH FAIL — expected food write missing — local foodEntries delta is 0 for this turn (proven no food write, not a capture gap)

Why verdict WRITE-TRUTH FAIL — expected food write missing — local foodEntries delta is 0 for this turn (proven no food write, not a capture gap)

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 no speech captured

2 · Card shown Logged your note in General about two chicken breakfast sausages.

4 · UI did navigated to "notes" 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 2026-07-30T20:27:21.351Z

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 food"Supper was one cup kidney beans." (difficulty —)3.0s

Verdict Expected LOG — should log the food. 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 3.0s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for one cup kidney beans? 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 2026-07-30T20:27:50.833Z

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 food"Enter one medium apricot." (difficulty —)2.5s

Verdict Expected LOG — should log the food. 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 2.5s

1 · TTS said no speech captured

2 · Card shown I could not finish that request safely.

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 2026-07-30T20:28:04.534Z

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 food"Ate a third cup dried mango." (difficulty —)0.7s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Mango, Dried — card not captured.

Why verdict Logged (write-truth verified): Mango, Dried — 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 no speech captured

2 · Card shown Logged Mango, Dried (a third cup (55 g)).

3 · App data rows written created food_log_entry: Mango, Dried ×1 (a third cup (55 g)) 175 cal · 1.4g P · 43.2g 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 2026-07-30T20:28:16.434Z

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

7 · Lookup method The answer used provisional food evidence from the USDA nutrition database.

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

FAILshould log the food"Track one tablespoon cashew butter." (difficulty —)0.2s

Verdict Expected LOG — should log the food. FAIL: WRITE-TRUTH FAIL — MACROS off for "cashew butter": logged 180 kcal vs expected ~94 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "cashew butter": logged 180 kcal vs expected ~94 (§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 no speech captured

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

3 · App data rows written created food_log_entry: cashew butter ×1 (serving) 180 cal · 8g P · 18g 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 2026-07-30T20:28:27.750Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Dinner side was one cup roasted beets." (difficulty —)0.3s

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

Why verdict Logged (write-truth verified): Beets — 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 no speech captured

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

3 · App data rows written created food_log_entry: Beets ×1 (1 cup) 75 cal · 2.9g P · 17g 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 2026-07-30T20:28:39.184Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I had two cups pineapple chunks." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Pineapple — 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 no speech captured

2 · Card shown Logged two cups pineapple chunks.

3 · App data rows written created food_log_entry: Pineapple ×1 (2 cup) 165 cal · 1.7g P · 43.2g 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 2026-07-30T20:28:50.500Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Put in one hundred eighty grams skyr-style yogurt." (difficulty —)0.2s

Verdict Expected LOG — should log the food. 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 0.2s

1 · TTS said no speech captured

2 · Card shown I found Plain Greek yogurt, but I also heard hundred eighty skyr style 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 2026-07-30T20:29:01.851Z

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 food"Log one cup almond milk unsweetened." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Unsweetened almond 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 no speech captured

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

3 · App data rows written created food_log_entry: Unsweetened almond milk ×1 (1 cup) 31 cal · 1g P · 0.7g C · 2.6g 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 2026-07-30T20:29:13.167Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Breakfast was half a cup feta crumbles." (difficulty —)0.3s

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

Why verdict Logged (write-truth verified): 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 0.3s

1 · TTS said no speech captured

2 · Card shown Logged half a cup feta crumbles. 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: Feta cheese ×1 (0.5 cup) 198 cal · 10.6g P · 3.1g C · 16g 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 2026-07-30T20:29:24.634Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Count two corn tortillas small." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Corn 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 no speech captured

2 · Card shown Logged two corn tortillas small.

3 · App data rows written created food_log_entry: Corn tortilla ×1 (2 corn tortillas) 122 cal · 3.2g P · 25g 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 2026-07-30T20:29:35.982Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I had six ounces lean ground chicken." (difficulty —)0.2s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Lean ground chicken — card not captured.

Why verdict Logged (write-truth verified): Lean ground chicken — 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 no speech captured

2 · Card shown Logged six ounces lean ground chicken.

3 · App data rows written created food_log_entry: Lean ground chicken ×1 (6 oz) 191 cal · 32.3g 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 2026-07-30T20:29:47.284Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Add one cup black-eyed peas." (difficulty —)5.4s

Verdict Expected LOG — should log the food. 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 no speech captured

2 · Card shown What exact food and amount should I use for Cooked peas? 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 2026-07-30T20:30:03.768Z

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 food"Track half a cup pico de gallo." (difficulty —)0.7s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Salsa, Pico De Gallo — card not captured.

Why verdict Logged (write-truth verified): Salsa, Pico De Gallo — 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 no speech captured

2 · Card shown I logged Salsa, Pico De Gallo (half a cup (120 g)) — let me know if that's not right.

3 · App data rows written created food_log_entry: Salsa, Pico De Gallo ×1 (half a cup (120 g)) 20 cal · 0.8g P · 4.4g 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 2026-07-30T20:30:15.568Z

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

7 · Lookup method The answer used provisional food evidence from the USDA nutrition database.

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

FAILshould log the food"Log two frozen waffles whole grain." (difficulty —)2.1s

Verdict Expected LOG — should log the food. 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 2.1s

1 · TTS said no speech captured

2 · Card shown Did you have syrup, butter, or another topping with it? 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 2026-07-30T20:30:28.852Z

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 food"I ate one plain croissant small." (difficulty —)2.7s

Verdict Expected LOG — should log the food. 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 2.7s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for one plain croissant small? 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 2026-07-30T20:30:42.701Z

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 food"Record five ounces mussels." (difficulty —)0.3s

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

Why verdict Logged (write-truth verified): Mussels — 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 no speech captured

2 · Card shown I logged Mussels (five ounces (141.7 g)) — let me know if that's not right.

3 · App data rows written created food_log_entry: Mussels ×1 (five ounces (141.7 g)) 155 cal · 21.4g P · 6.7g 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 2026-07-30T20:30:54.267Z

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

7 · Lookup method The answer used provisional food evidence from the USDA nutrition database.

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

PASSshould log the food"Snack was one Babybel light." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Babybel light — 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 no speech captured

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

3 · App data rows written created food_log_entry: Babybel light ×1 (1 babybel light) 42 cal · 6g P · 0g C · 2.5g 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 2026-07-30T20:31:05.567Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Add one teaspoon avocado oil." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Avocado oil — 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 no speech captured

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

3 · App data rows written created food_log_entry: Avocado oil ×1 (1 tsp) 40 cal · 0g P · 0g 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 2026-07-30T20:31:16.868Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I had one cup barley." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Cooked barley — 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 no speech captured

2 · Card shown Logged one cup barley. 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 barley ×1 (1 cup) 193 cal · 3.6g P · 44.3g 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 2026-07-30T20:31:28.168Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Log one medium sweet onion sauteed." (difficulty —)2.7s

Verdict Expected LOG — should log the food. 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 2.7s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for one medium sweet onion sauteed? 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 2026-07-30T20:31:42.001Z

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 food"Track eight ounces cold brew coffee." (difficulty —)5.3s

Verdict Expected LOG — should log the food. 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 5.3s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for Coffee? 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 2026-07-30T20:32:13.918Z

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 food"Count one cup radishes." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Radishes — 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 no speech captured

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

3 · App data rows written created food_log_entry: Radishes ×1 (1 cup) 19 cal · 0.8g P · 3.9g 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 2026-07-30T20:32:25.217Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Dinner was six ounces turkey meatballs." (difficulty —)0.2s

Verdict Expected LOG — should log the food. FAIL: WRITE-TRUTH FAIL — MACROS off for "turkey meatballs": logged 315 kcal vs expected ~220 (§14 15pct/25cal band)

Why verdict WRITE-TRUTH FAIL — MACROS off for "turkey meatballs": logged 315 kcal vs expected ~220 (§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 no speech captured

2 · Card shown Logged six ounces turkey meatballs.

3 · App data rows written created food_log_entry: Turkey meatballs ×1 (6 oz) 315 cal · 30.6g P · 10.2g C · 17g 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 2026-07-30T20:32:36.518Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I had one cup miso soup." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Miso soup — 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 no speech captured

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

3 · App data rows written created food_log_entry: Miso soup ×1 (1 cup) 49 cal · 3.7g P · 6.6g C · 1.5g 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 2026-07-30T20:32:47.820Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Add three rice crackers." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Rice crackers — 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 no speech captured

2 · Card shown Logged three rice crackers.

3 · App data rows written created food_log_entry: Rice crackers ×1 (3 rice crackers) 87 cal · 2.1g P · 17.3g C · 1.1g 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 2026-07-30T20:32:59.102Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Log one cup blueberries frozen." (difficulty —)2.3s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Blueberries, Frozen — card not captured.

Why verdict Logged (write-truth verified): Blueberries, Frozen — 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 no speech captured

2 · Card shown I logged Blueberries, Frozen (1 serving) (one cup (148 g)) — let me know if that's not right.

3 · App data rows written created food_log_entry: Blueberries, Frozen ×1 (one cup (148 g)) 75 cal · 0.6g P · 18.1g 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 2026-07-30T20:33:12.550Z

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

7 · Lookup method The answer used provisional food evidence from a manually curated nutrition entry.

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

PASSshould log the food"Track two tablespoons tzatziki." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Tzatziki — 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 no speech captured

2 · Card shown Logged two tablespoons tzatziki.

3 · App data rows written created food_log_entry: Tzatziki ×1 (2 tbsp) 28 cal · 1.1g P · 1.3g C · 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 2026-07-30T20:33:23.870Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I ate one medium guava." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Guava — 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 no speech captured

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

3 · App data rows written created food_log_entry: Guava ×1 (1 guava) 37 cal · 1.4g P · 7.9g 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 2026-07-30T20:33:35.135Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Put down one cup bulgur." (difficulty —)2.7s

Verdict Expected LOG — should log the food. 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 2.7s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for one cup bulgur? 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 2026-07-30T20:33:48.968Z

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 food"Record four ounces chicken breast strips." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Chicken breast — 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 no speech captured

2 · Card shown Logged four ounces chicken breast strips.

3 · App data rows written created food_log_entry: Chicken breast ×1 (4 oz) 187 cal · 35.2g P · 0g C · 4.1g 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 2026-07-30T20:34:00.268Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Add my usual Aloha bar." (difficulty —)2.0s

Verdict Expected LOG — should log the food. 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 2.0s

1 · TTS said no speech captured

2 · Card shown Which one did you mean? 1. my usual Aloha bar, 2. Aloha Protein Bar, 3. Protein Bar — Aloha

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Aloha Aloha Chocolate Chip Cookie Dough

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-aloha",
    "name": "Aloha Chocolate Chip Cookie Dough",
    "brand": "Aloha",
    "source": "manual",
    "servingLabel": "bar",
    "servingQuantity": 1,
    "servingUnit": "bar",
    "macrosPerServing": {
      "calories": 220,
      "proteinGrams": 14,
      "fatGrams": 10,
      "carbsGrams": 24
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual Aloha bar",
  "foodItemId": "seed-f-aloha",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "bar",
    "displayLabel": "my usual Aloha bar",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:34:13.402Z

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 food"I had my usual labneh." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Plain Labneh — 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 no speech captured

2 · Card shown Logged Codex Test Dairy Plain Labneh. Using your saved "my usual labneh". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Plain Labneh ×1 (cup) 160 cal · 10g P · 5g C · 12g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Dairy Plain Labneh

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-labneh",
    "name": "Plain Labneh",
    "brand": "Codex Test Dairy",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 160,
      "proteinGrams": 10,
      "fatGrams": 12,
      "carbsGrams": 5
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual labneh",
  "foodItemId": "seed-f-labneh",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual labneh",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:34:24.701Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Lunch was my usual chicken shawarma." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Chicken Shawarma Plate — 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 no speech captured

2 · Card shown Logged Codex Test Deli Chicken Shawarma Plate. Using your saved "my usual chicken shawarma". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Chicken Shawarma Plate ×1 (plate) 480 cal · 36g P · 40g C · 18g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Deli Chicken Shawarma Plate

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-shawarma",
    "name": "Chicken Shawarma Plate",
    "brand": "Codex Test Deli",
    "source": "manual",
    "servingLabel": "plate",
    "servingQuantity": 1,
    "servingUnit": "plate",
    "macrosPerServing": {
      "calories": 480,
      "proteinGrams": 36,
      "fatGrams": 18,
      "carbsGrams": 40
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual chicken shawarma",
  "foodItemId": "seed-f-shawarma",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "plate",
    "displayLabel": "my usual chicken shawarma",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:34:35.899Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Dinner side was my usual rice." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Cooked Jasmine 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 0.1s

1 · TTS said no speech captured

2 · Card shown Logged Cooked Jasmine Rice. Using your saved "my usual rice". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Cooked Jasmine Rice ×1 (cup) 205 cal · 4g P · 45g C · 0.4g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Cooked Jasmine Rice

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-rice",
    "name": "Cooked Jasmine Rice",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 205,
      "proteinGrams": 4,
      "fatGrams": 0.4,
      "carbsGrams": 45
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual rice",
  "foodItemId": "seed-f-rice",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual rice",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:34:47.206Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Track my usual pistachios." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Pistachios — 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 no speech captured

2 · Card shown Logged Pistachios. Using your saved "my usual pistachios". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Pistachios ×1 (ounce) 160 cal · 6g P · 8g C · 13g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Pistachios

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-pistachio",
    "name": "Pistachios",
    "brand": null,
    "source": "manual",
    "servingLabel": "ounce",
    "servingQuantity": 1,
    "servingUnit": "oz",
    "macrosPerServing": {
      "calories": 160,
      "proteinGrams": 6,
      "fatGrams": 13,
      "carbsGrams": 8
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual pistachios",
  "foodItemId": "seed-f-pistachio",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "ounce",
    "displayLabel": "my usual pistachios",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:34:58.415Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Log my default almond milk." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Unsweetened Almond 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 no speech captured

2 · Card shown Logged Unsweetened Almond Milk. Using your saved "my default almond milk". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Unsweetened Almond Milk ×1 (cup) 30 cal · 1g P · 1g C · 2.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Unsweetened Almond Milk

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-almondmilk",
    "name": "Unsweetened Almond Milk",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 30,
      "proteinGrams": 1,
      "fatGrams": 2.5,
      "carbsGrams": 1
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my default almond milk",
  "foodItemId": "seed-f-almondmilk",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my default almond milk",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:35:09.645Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Add my usual evening chai." (difficulty —)0.1s

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

Why verdict Logged (write-truth verified): Chai 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 no speech captured

2 · Card shown Logged Codex Test Cafe Chai Latte. Using your saved "my usual evening chai". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Chai Latte ×1 (cup) 120 cal · 4g P · 20g C · 3g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Cafe Chai Latte

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-chai",
    "name": "Chai Latte",
    "brand": "Codex Test Cafe",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 120,
      "proteinGrams": 4,
      "fatGrams": 3,
      "carbsGrams": 20
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual evening chai",
  "foodItemId": "seed-f-chai",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual evening chai",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:35:20.860Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 grits." (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 no speech captured

2 · Card shown Logged Cooked Grits. Using your saved "my normal grits". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Cooked Grits ×1 (cup) 145 cal · 3g P · 31g C · 0.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Cooked Grits

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-grits",
    "name": "Cooked Grits",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 145,
      "proteinGrams": 3,
      "fatGrams": 0.5,
      "carbsGrams": 31
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my normal grits",
  "foodItemId": "seed-f-grits",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my normal grits",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:35:32.075Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Track my go-to cold brew." (difficulty —)0.1s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Cold Brew Coffee — card not captured.

Why verdict Logged (write-truth verified): Cold Brew Coffee — 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 no speech captured

2 · Card shown Logged Codex Test Cafe Cold Brew Coffee. Using your saved "my go to cold brew". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Cold Brew Coffee ×1 (cup) 5 cal · 0.3g P · 0g C · 0g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Cafe Cold Brew Coffee

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-coldbrew",
    "name": "Cold Brew Coffee",
    "brand": "Codex Test Cafe",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 5,
      "proteinGrams": 0.3,
      "fatGrams": 0,
      "carbsGrams": 0
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my go-to cold brew",
  "foodItemId": "seed-f-coldbrew",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my go-to cold brew",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:35:43.309Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Log my usual bedtime cocoa." (difficulty —)0.1s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Hot Cocoa Water — card not captured.

Why verdict Logged (write-truth verified): Hot Cocoa Water — 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 no speech captured

2 · Card shown Logged Hot Cocoa Water. Using your saved "my usual bedtime cocoa". Tell me if that is wrong.

3 · App data rows written created food_log_entry: Hot Cocoa Water ×1 (cup) 80 cal · 2g P · 16g C · 1g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Hot Cocoa Water

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-cocoa",
    "name": "Hot Cocoa Water",
    "brand": null,
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 80,
      "proteinGrams": 2,
      "fatGrams": 1,
      "carbsGrams": 16
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

food_default

{
  "type": "food_default",
  "phrase": "my usual bedtime cocoa",
  "foodItemId": "seed-f-cocoa",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "cup",
    "displayLabel": "my usual bedtime cocoa",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:35:54.607Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Add the same feta as yesterday." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): 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 0.2s

1 · TTS said no speech captured

2 · Card shown Logged Feta cheese. Using your recent Feta cheese history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Feta cheese ×1 (0.5 cup) 198 cal · 10.6g P · 3.1g C · 16g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Dairy Feta Crumbles

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-feta",
    "name": "Feta Crumbles",
    "brand": "Codex Test Dairy",
    "source": "manual",
    "servingLabel": "cup",
    "servingQuantity": 1,
    "servingUnit": "cup",
    "macrosPerServing": {
      "calories": 200,
      "proteinGrams": 10,
      "fatGrams": 16,
      "carbsGrams": 3
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 0.5 for seed-f-hist-feta at 2026-07-17T13:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-feta",
  "quantity": 0.5,
  "loggedAt": "2026-07-17T13:00:00.000Z"
}

snapshot captured 2026-07-30T20:36:21.258Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

UNVERIFIEDunclassifiedshould log the food"Same Aloha bar as yesterday." (difficulty —)3.2s

Verdict Expected LOG — should log the food. UNVERIFIED: CONTEXT-FRAGMENT — app asked for missing context; needs multi-turn replay (not a single-shot fail).

Why verdict CONTEXT-FRAGMENT — app asked for missing context; needs multi-turn replay (not a single-shot fail).

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 3.2s

1 · TTS said no speech captured

2 · Card shown I couldn't find Same Aloha Snack Bar in my nutrition database, so I haven't logged it — I'd only be guessing the nutrition. Say the exact name on the package, or tell me to log a generic snack bar.

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Aloha Aloha Protein Bar

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-aloha",
    "name": "Aloha Protein Bar",
    "brand": "Aloha",
    "source": "manual",
    "servingLabel": "bar",
    "servingQuantity": 1,
    "servingUnit": "bar",
    "macrosPerServing": {
      "calories": 220,
      "proteinGrams": 14,
      "fatGrams": 10,
      "carbsGrams": 24
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-f-hist-aloha at 2026-07-17T16:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-aloha",
  "quantity": 1,
  "loggedAt": "2026-07-17T16:00:00.000Z"
}

snapshot captured 2026-07-30T20:36:35.606Z

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 food"Log the same miso soup I had last night." (difficulty —)0.3s

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

Why verdict Logged (write-truth verified): Miso soup — 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 no speech captured

2 · Card shown Logged Miso soup. Using your recent Miso soup history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Miso soup ×1 (1 cup) 49 cal · 3.7g P · 6.6g C · 1.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Codex Test Kitchen Miso Soup Bowl

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-miso",
    "name": "Miso Soup Bowl",
    "brand": "Codex Test Kitchen",
    "source": "manual",
    "servingLabel": "bowl",
    "servingQuantity": 1,
    "servingUnit": "bowl",
    "macrosPerServing": {
      "calories": 50,
      "proteinGrams": 3,
      "fatGrams": 1,
      "carbsGrams": 6
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-f-hist-miso at 2026-07-17T19:30:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-miso",
  "quantity": 1,
  "loggedAt": "2026-07-17T19:30:00.000Z"
}

snapshot captured 2026-07-30T20:36:47.058Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Same oatmeal as this morning." (difficulty —)0.2s

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

Why verdict Logged (write-truth verified): Oatmeal Bowl — 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 no speech captured

2 · Card shown Logged Oatmeal Bowl. Using your recent Oatmeal Bowl history. Tell me if that is wrong.

3 · App data rows written created food_log_entry: Oatmeal Bowl ×1 (bowl) 150 cal · 5g P · 27g C · 2.5g F

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Oatmeal Bowl

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-oats",
    "name": "Oatmeal Bowl",
    "brand": null,
    "source": "manual",
    "servingLabel": "bowl",
    "servingQuantity": 1,
    "servingUnit": "bowl",
    "macrosPerServing": {
      "calories": 150,
      "proteinGrams": 5,
      "fatGrams": 2.5,
      "carbsGrams": 27
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-f-hist-oats at 2026-07-18T07:30:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-oats",
  "quantity": 1,
  "loggedAt": "2026-07-18T07:30:00.000Z"
}

snapshot captured 2026-07-30T20:36:58.393Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

FAILshould log the food"Add the same ahi I had earlier." (difficulty —)0.3s

Verdict Expected LOG — should log the food. 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 no speech captured

2 · Card shown I hear more than one saved version that could match "the same ahi i had earlier". Which exact food and amount should I log?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Seared Ahi Tuna

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-ahi",
    "name": "Seared Ahi Tuna",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 180,
      "proteinGrams": 38,
      "fatGrams": 2,
      "carbsGrams": 0
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-f-hist-ahi at 2026-07-18T12:30:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-ahi",
  "quantity": 1,
  "loggedAt": "2026-07-18T12:30:00.000Z"
}

snapshot captured 2026-07-30T20:37:09.841Z

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 log the food"Same smoothie as yesterday." (difficulty —)0.2s

Verdict Expected LOG — should log the food. UNVERIFIED: CONTEXT-FRAGMENT — app asked for missing context; needs multi-turn replay (not a single-shot fail).

Why verdict CONTEXT-FRAGMENT — app asked for missing context; needs multi-turn replay (not a single-shot fail).

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 no speech captured

2 · Card shown I hear more than one saved version that could match "same smoothie as yesterday". 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 Recipe Cherry Protein Smoothie

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-smooth",
    "name": "Cherry Protein Smoothie",
    "brand": "Codex Test Recipe",
    "source": "manual",
    "servingLabel": "smoothie",
    "servingQuantity": 1,
    "servingUnit": "smoothie",
    "macrosPerServing": {
      "calories": 270,
      "proteinGrams": 22,
      "fatGrams": 4,
      "carbsGrams": 32
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-f-hist-smooth at 2026-07-17T10:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-smooth",
  "quantity": 1,
  "loggedAt": "2026-07-17T10:00:00.000Z"
}

snapshot captured 2026-07-30T20:37:21.240Z

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 food"Log the shrimp I had last time." (difficulty —)0.1s

Verdict Expected LOG — should log the food. 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 0.1s

1 · TTS said no speech captured

2 · Card shown I heard you want the same the shrimp i had last time 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: Grilled Shrimp

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-shrimp",
    "name": "Grilled Shrimp",
    "brand": null,
    "source": "manual",
    "servingLabel": "serving",
    "servingQuantity": 1,
    "servingUnit": "serving",
    "macrosPerServing": {
      "calories": 170,
      "proteinGrams": 36,
      "fatGrams": 2,
      "carbsGrams": 1
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 1 for seed-f-hist-shrimp at 2026-07-16T18:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-shrimp",
  "quantity": 1,
  "loggedAt": "2026-07-16T18:00:00.000Z"
}

snapshot captured 2026-07-30T20:37:32.458Z

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 log the food"Same tortilla as earlier today." (difficulty —)0.4s

Verdict Expected LOG — should log the food. UNVERIFIED: CONTEXT-FRAGMENT — app asked for missing context; needs multi-turn replay (not a single-shot fail).

Why verdict CONTEXT-FRAGMENT — app asked for missing context; needs multi-turn replay (not a single-shot fail).

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 no speech captured

2 · Card shown I hear more than one saved version that could match "same tortilla as earlier today". Which exact food and amount should I log?

4 · UI did no screen events captured

5 · Seed/context data

Seeded starting state:

Seed food: Corn Tortilla

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-hist-tort",
    "name": "Corn Tortilla",
    "brand": null,
    "source": "manual",
    "servingLabel": "tortilla",
    "servingQuantity": 1,
    "servingUnit": "tortilla",
    "macrosPerServing": {
      "calories": 50,
      "proteinGrams": 1.2,
      "fatGrams": 0.7,
      "carbsGrams": 10
    },
    "aliases": [],
    "timesLogged": 0,
    "createdAt": "2026-07-16T12:00:00.000Z"
  }
}

Seed food history: 2 for seed-f-hist-tort at 2026-07-18T07:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-hist-tort",
  "quantity": 2,
  "loggedAt": "2026-07-18T07:00:00.000Z"
}

snapshot captured 2026-07-30T20:37:43.990Z

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 food"Snack was one Larabar cashew cookie." (difficulty —)4.9s

Verdict Expected LOG — should log the food. 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 4.9s

1 · TTS said no speech captured

2 · Card shown I could not finish that request safely.

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 2026-07-30T20:38:00.040Z

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 food"Finished one Fairlife Core Power banana shake." (difficulty —)0.1s

Why verdict WRITE-TRUTH FAIL — expected food write missing — local foodEntries delta is 0 for this turn (proven no food write, not a capture gap)

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 speech captured

2 · Card shown I could not find a to-do matching one Fairlife Core Power banana shake.

4 · UI did navigated to "todos" 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 2026-07-30T20:38:11.224Z

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 food"Track one Fage total zero strawberry cup." (difficulty —)7.3s

Verdict Expected LOG — should log the food. 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 7.3s

1 · TTS said no speech captured

2 · Card shown Which one did you mean? 1. Total 2% Milkfat Yogurt — FAGE, 2. Total 0% Milkfat — FAGE, 3. Total 5% Milkfat — FAGE

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 2026-07-30T20:38:29.689Z

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 food"Save one Aloha chocolate mint bar." (difficulty —)0.5s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Chocolate Mint Organic Protein Bar — card not captured.

Why verdict Logged (write-truth verified): Chocolate Mint Organic 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 no speech captured

2 · Card shown Logged Chocolate Mint Organic Protein Bar — ALOHA — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Chocolate Mint Organic Protein Bar ×1 (1 Bar) 220 cal · 14g P · 26g C · 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 2026-07-30T20:38:41.291Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"I ate one RXBAR mixed berry." (difficulty —)0.4s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Mixed Berry Protein Bar — card not captured.

Why verdict Logged (write-truth verified): Mixed Berry 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 0.4s

1 · TTS said no speech captured

2 · Card shown Logged Mixed Berry Protein Bar — RXBAR — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Mixed Berry Protein Bar ×1 (1 bar) 210 cal · 12g P · 24g 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 2026-07-30T20:38:52.840Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould log the food"Log a papaya and one Aloha bar." (difficulty —)0.6s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Aloha Protein Bar, Chocolate Espresso, Papaya — card not captured.

Why verdict Logged (write-truth verified): Aloha Protein Bar, Chocolate Espresso, Papaya — 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 no speech captured

2 · Card shown Logged a papaya and Aloha Protein Bar, Chocolate Espresso — Aloha. 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: Papaya ×1 (100 g) 43 cal · 0.5g P · 10.8g C · 0.3g F
created food_log_entry: Aloha Protein Bar, Chocolate Espresso ×1 (1 serving (56 g)) 230 cal · 14g P · 26g C · 10g 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 2026-07-30T20:39:04.557Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I had oatmeal and cold brew." (difficulty —)0.5s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Cooked oatmeal, cold brew — card not captured.

Why verdict Logged (write-truth verified): Cooked oatmeal, cold brew — 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 no speech captured

2 · Card shown Logged Cooked oatmeal and cold brew. 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 oatmeal ×1 (100 g) 71 cal · 2.5g P · 12g C · 1.5g F
created food_log_entry: cold brew ×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 2026-07-30T20:39:16.157Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Breakfast was sunny side eggs and corn tortillas." (difficulty —)0.4s

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

Why verdict Logged (write-truth verified): Corn tortilla, 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 no speech captured

2 · Card shown Logged Egg and Corn tortilla. 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: Egg ×1 (50 g) 72 cal · 6.3g P · 0.4g C · 4.8g F
created food_log_entry: Corn tortilla ×1 (28 g) 61 cal · 1.6g P · 12.5g 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 2026-07-30T20:39:27.655Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Lunch was a turkey burger and a tangerine." (difficulty —)5.1s

Verdict Expected LOG — should log the food. 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 5.1s

1 · TTS said no speech captured

2 · Card shown I could not finish that request safely.

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 2026-07-30T20:39:43.957Z

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 food"Snack was pistachios and cherries." (difficulty —)0.2s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Cherries, Pistachios — card not captured.

Why verdict Logged (write-truth verified): Cherries, Pistachios — 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 no speech captured

2 · Card shown Logged Pistachios and Cherries. 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: Pistachios ×1 (1 oz) 157 cal · 5.7g P · 7.6g C · 12.7g F
created food_log_entry: Cherries ×1 (154 g) 97 cal · 1.7g P · 24.6g 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 2026-07-30T20:39:55.306Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Dinner was ahi and roasted carrots." (difficulty —)0.2s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Baby carrots, Ahi tuna — card not captured.

Why verdict Logged (write-truth verified): Baby carrots, Ahi tuna — 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 no speech captured

2 · Card shown Logged Ahi tuna and Baby carrots. 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: Ahi tuna ×1 (85 g) 111 cal · 24.7g P · 0g C · 0.5g F
created food_log_entry: Baby carrots ×1 (10 g) 4 cal · 0.1g P · 0.8g 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 2026-07-30T20:40:06.622Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I ate labneh and blueberries." (difficulty —)0.2s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Labneh, Blueberries — card not captured.

Why verdict Logged (write-truth verified): Labneh, Blueberries — 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 no speech captured

2 · Card shown Logged Labneh and Blueberries. 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: Labneh ×1 (100 g) 80 cal · 5g P · 2.5g C · 6g F
created food_log_entry: Blueberries ×1 (148 g) 84 cal · 1g P · 21.5g 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 2026-07-30T20:40:17.957Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Track shrimp and jasmine rice." (difficulty —)0.3s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Cooked jasmine rice, Shrimp — card not captured.

Why verdict Logged (write-truth verified): Cooked jasmine rice, Shrimp — 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 no speech captured

2 · Card shown Logged Shrimp and Cooked jasmine rice. 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: Shrimp ×1 (1 oz) 28 cal · 6.7g P · 0.1g C · 0.1g F
created food_log_entry: Cooked jasmine rice ×1 (100 g) 130 cal · 2.4g P · 28.2g 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 2026-07-30T20:40:40.924Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Add feta and jicama." (difficulty —)0.3s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Jicama, Feta cheese — card not captured.

Why verdict Logged (write-truth verified): Jicama, 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 0.3s

1 · TTS said no speech captured

2 · Card shown Logged Feta cheese and Jicama. 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: Feta cheese ×1 (100 g) 264 cal · 14.2g P · 4.1g C · 21.3g F
created food_log_entry: Jicama ×1 (120 g) 46 cal · 0.8g P · 10.6g 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 2026-07-30T20:40:52.322Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I had a Larabar and sparkling lime water." (difficulty —)0.7s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Larabar, Water — card not captured.

Why verdict Logged (write-truth verified): Larabar, Water — 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 no speech captured

2 · Card shown Logged Larabar — Larabar 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: Larabar ×1 (55 g) 200 cal · 20g P · 22g C · 7g 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 2026-07-30T20:41:04.307Z

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

7 · Lookup method The answer used provisional food evidence from a manually curated nutrition entry.

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

PASSshould log the food"Log miso soup and bok choy." (difficulty —)0.3s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Miso soup, Bok choy — card not captured.

Why verdict Logged (write-truth verified): Miso soup, Bok choy — 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 no speech captured

2 · Card shown Logged Miso soup and Bok choy. 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: Miso soup ×1 (100 g) 20 cal · 1.5g P · 2.7g C · 0.6g F
created food_log_entry: Bok choy ×1 (70 g) 9 cal · 1g P · 1.5g 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 2026-07-30T20:41:15.707Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Breakfast was oatmeal, yogurt, and coffee." (difficulty —)2.4s

Verdict Expected LOG — should log the food. 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 2.4s

1 · TTS said no speech captured

2 · Card shown I could not finish that request safely.

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 2026-07-30T20:41:29.206Z

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 food"I ate a guava, Babybel light, and rice crackers." (difficulty —)0.3s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Rice crackers, Babybel light, Guava — card not captured.

Why verdict Logged (write-truth verified): Rice crackers, Babybel light, Guava — 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 no speech captured

2 · Card shown Logged a guava, Babybel light, and Rice crackers. 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: Guava ×1 (1 guava) 37 cal · 1.4g P · 7.9g C · 0.6g F
created food_log_entry: Babybel light ×1 (20 g) 42 cal · 6g P · 0g C · 2.5g F
created food_log_entry: Rice crackers ×1 (30 g) 125 cal · 3g P · 24.8g C · 1.5g 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 2026-07-30T20:41:40.623Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Dinner was turkey meatballs, barley, and beets." (difficulty —)0.2s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Turkey meatballs, Beets, Cooked barley — card not captured.

Why verdict Logged (write-truth verified): Turkey meatballs, Beets, Cooked barley — 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 no speech captured

2 · Card shown Logged Turkey meatballs, Cooked barley, and Beets. 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: Turkey meatballs ×1 (28 g) 52 cal · 5g P · 1.7g C · 2.8g F
created food_log_entry: Cooked barley ×1 (100 g) 123 cal · 2.3g P · 28.2g C · 0.4g F
created food_log_entry: Beets ×1 (85 g) 37 cal · 1.4g P · 8.5g 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 2026-07-30T20:41:52.056Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Snack was radishes and tzatziki." (difficulty —)0.2s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Radishes, Tzatziki — card not captured.

Why verdict Logged (write-truth verified): Radishes, Tzatziki — 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 no speech captured

2 · Card shown Logged Radishes and Tzatziki. 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: Radishes ×1 (116 g) 19 cal · 0.8g P · 3.9g C · 0.1g F
created food_log_entry: Tzatziki ×1 (30 g) 28 cal · 1.1g P · 1.3g C · 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 2026-07-30T20:42:03.371Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Lunch was a grain bowl and iced tea." (difficulty —)1.5s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Unsweetened iced tea, Grain Bowl — card not captured.

Why verdict Logged (write-truth verified): Unsweetened iced tea, Grain Bowl — 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 1.5s

1 · TTS said no speech captured

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 2026-07-30T20:42:15.973Z

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

7 · Lookup method The answer used provisional food evidence from a manually curated nutrition entry.

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

PASSshould log the food"I had rice, kidney beans, and ground chicken." (difficulty —)0.4s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Cooked white rice, Beans, Lean ground chicken — card not captured.

Why verdict Logged (write-truth verified): Cooked white rice, Beans, Lean ground chicken — 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 no speech captured

2 · Card shown Logged Cooked white rice, Beans, and Lean ground chicken. 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 white rice ×1 (100 g) 130 cal · 2.7g P · 28g C · 0.3g F
created food_log_entry: Beans ×1 (86 g) 109 cal · 7.5g P · 19.6g C · 0.4g F
created food_log_entry: Lean ground chicken ×1 (100 g) 112 cal · 19g P · 0g 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 2026-07-30T20:42:27.639Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Track labneh, honey, and pistachios." (difficulty —)0.4s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Pistachios, Labneh, Honey — card not captured.

Why verdict Logged (write-truth verified): Pistachios, Labneh, Honey — 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 no speech captured

2 · Card shown Logged Labneh, Honey, and Pistachios. 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: Labneh ×1 (100 g) 80 cal · 5g P · 2.5g C · 6g F
created food_log_entry: Honey ×1 (21 g) 64 cal · 0.1g P · 17.3g C · 0g F
created food_log_entry: Pistachios ×1 (1 oz) 157 cal · 5.7g P · 7.6g C · 12.7g 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 2026-07-30T20:42:39.122Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Add eggs, chicken sausage, and pineapple juice." (difficulty —)0.9s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Egg, AMYLU Chicken Sausage, Pineapple Juice, 100% — card not captured.

Why verdict Logged (write-truth verified): Egg, AMYLU Chicken Sausage, Pineapple 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 0.9s

1 · TTS said no speech captured

2 · Card shown Logged two eggs, AMYLU Chicken Sausage, and Pineapple Juice, 100%. Assumed a typical serving of AMYLU Chicken Sausage while logging the rest of this meal. Tell me if that is not right.

3 · App data rows written created food_log_entry: Egg ×1 (2 eggs) 143 cal · 12.6g P · 0.7g C · 9.5g F
created food_log_entry: AMYLU Chicken Sausage ×1 (one) 110 cal · 12g P · 1g C · 6g F
created food_log_entry: Pineapple Juice, 100% ×1 (1 fl oz (no ice)) 16 cal · 0.1g P · 4g 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 2026-07-30T20:42:51.105Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"I ate pasta and a side of butter lettuce." (difficulty —)0.4s

Verdict Expected LOG — should log the food. PASS: Logged (write-truth verified): Butter lettuce, Cooked pasta — card not captured.

Why verdict Logged (write-truth verified): Butter lettuce, Cooked pasta — 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 no speech captured

2 · Card shown Logged Cooked pasta and Butter lettuce. 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: Butter lettuce ×1 (55 g) 7 cal · 0.8g P · 1.2g 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 2026-07-30T20:43:02.623Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

PASSshould log the food"Log a kiwi and cashew butter." (difficulty —)0.5s

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

Why verdict Logged (write-truth verified): Kiwi, 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 0.5s

1 · TTS said no speech captured

2 · Card shown Logged a kiwi and Butter. 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: Kiwi ×1 (1 kiwi) 42 cal · 0.8g P · 10.1g C · 0.3g F
created food_log_entry: Butter ×1 (5 g) 36 cal · 0g P · 0g C · 4.1g 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 2026-07-30T20:43:14.206Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould log the food"Breakfast was cereal and almond milk." (difficulty —)4.9s

Verdict Expected LOG — should log the food. 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 4.9s

1 · TTS said no speech captured

2 · Card shown I could not finish that request safely.

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 2026-07-30T20:43:30.224Z

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 food"I had huevos rancheros with two eggs, salsa, and black beans" (difficulty —)2.8s

Verdict Expected LOG — should log the food. 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 2.8s

1 · TTS said no speech captured

2 · Card shown Which salsa 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 2026-07-30T20:43:44.221Z

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 food"Log a smoothie with cherries, kale, and casein protein." (difficulty —)3.6s

Verdict Expected LOG — should log the food. FAIL: WRITE-TRUTH FAIL — COMPOUND: component name missing — expected like "casein", got: Cherries, Protein Bar, Kale, Raw

Why verdict WRITE-TRUTH FAIL — COMPOUND: component name missing — expected like "casein", got: Cherries, Protein Bar, Kale, Raw

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 no speech captured

2 · Card shown Logged Smoothie. Includes cherries, Kale, Raw (1 serving), and Aloha Protein Bar.

3 · App data rows written created food_log_entry: Smoothie ×1 (serving) 324 cal · 16.3g P · 49.5g C · 10.6g F
created food_log_entry: Cherries ×1 (154 g) 97 cal · 1.7g P · 24.6g C · 0.3g F
created food_log_entry: Kale, Raw ×1 (1 serving) 7 cal · 0.6g P · 0.9g C · 0.3g F
created food_log_entry: Protein Bar ×1 (bar) 220 cal · 14g P · 24g C · 10g 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 2026-07-30T20:43:58.972Z

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

7 · Lookup method The answer used provisional food evidence from a manually curated nutrition entry.

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

FAILshould log the food"I had a grain bowl with bulgur, lentils, and yogurt sauce." (difficulty —)11.3s

Verdict Expected LOG — should log the food. 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 11.3s

1 · TTS said no speech captured

2 · Card shown What should I use for yogurt sauce?

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 2026-07-30T20:44:21.439Z

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 food"Breakfast was breakfast burrito with egg, potato, and cheese." (difficulty —)2.3s

Verdict Expected LOG — should log the food. 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 2.3s

1 · TTS said no speech captured

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: none listed in the corpus row.

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

snapshot captured 2026-07-30T20:44:34.889Z

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 food"I made jambalaya with shrimp, sausage, rice, and peppers." (difficulty —)2.9s

Verdict Expected LOG — should log the food. 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 2.9s

1 · TTS said no speech captured

2 · Card shown What should I use for sausage?

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 2026-07-30T20:44:48.906Z

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 food"Dinner was a taco bowl with rice, carnitas, pico, and guacamole." (difficulty —)5.8s

Verdict Expected LOG — should log the food. 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 5.8s

1 · TTS said no speech captured

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 2026-07-30T20:45:21.256Z

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 log the food"I had ramen with chicken and corn." (difficulty —)2.6s

Verdict Expected LOG — should log the food. 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 no speech captured

2 · Card shown What should I use for corn?

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 2026-07-30T20:45:34.956Z

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 food"Log a chicken sandwich with pickles and mustard." (difficulty —)1.3s

Verdict Expected LOG — should log the food. 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 no speech captured

2 · Card shown What should I use for pickles?

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 2026-07-30T20:45:47.872Z

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 food"I built an acai bowl with granola, pineapple, and coconut." (difficulty —)4.1s

Verdict Expected LOG — should log the food. 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 4.1s

1 · TTS said no speech captured

2 · Card shown What should I use for coconut?

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 2026-07-30T20:46:03.138Z

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 food"Dinner was green curry with tofu, eggplant, and jasmine rice." (difficulty —)4.6s

Verdict Expected LOG — should log the food. 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 4.6s

1 · TTS said no speech captured

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 2026-07-30T20:46:18.956Z

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 some grits." (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 no speech captured

2 · Card shown How much should I log for Instant grits? 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 2026-07-30T20:46:30.205Z

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 bulgur." (difficulty —)0.7s

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 0.7s

1 · TTS said no speech captured

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

3 · App data rows written created food_log_entry: Bulgur, Cooked ×1 (1 serving) 151 cal · 5.6g P · 33.9g 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 2026-07-30T20:46:42.055Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 jasmine rice for dinner." (difficulty —)1.3s

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 1.3s

1 · TTS said no speech captured

2 · Card shown How much should I log for Cooked jasmine rice? 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 2026-07-30T20:46:54.555Z

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 ramen noodles." (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 no speech captured

2 · Card shown I logged Ramen Noodles — H & U, Inc. — let me know if that's not right.

3 · App data rows written created food_log_entry: Ramen Noodles ×1 (1 CONTAINER) 370 cal · 13g P · 76g 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 2026-07-30T20:47:06.170Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 tortilla chips." (difficulty —)0.3s

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 0.3s

1 · TTS said no speech captured

2 · Card shown Logged Tortilla Chips, Plain — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Tortilla Chips, Plain ×1 (1 small single serving bag) 132 cal · 2g P · 19g C · 5.8g 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 2026-07-30T20:47:17.605Z

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

7 · Lookup method The answer used provisional food evidence from the USDA nutrition database.

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 pistachios." (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 no speech captured

2 · Card shown Logged pistachios. Assumed about 1 oz (28 g) as a typical serving because no amount was stated — not one piece.

3 · App data rows written created food_log_entry: Pistachios ×1 (1 oz) 157 cal · 5.7g P · 7.6g C · 12.7g 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 2026-07-30T20:47:28.941Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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 miso." (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 no speech captured

2 · Card shown What exact food and amount should I use for a bowl of miso? 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 2026-07-30T20:47:42.788Z

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 jambalaya." (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 no speech captured

2 · Card shown What exact food and amount should I use for a plate of jambalaya? 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 2026-07-30T20:47:56.605Z

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 pineapple juice." (difficulty —)1.3s

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 1.3s

1 · TTS said no speech captured

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 2026-07-30T20:48:09.086Z

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 cold brew." (difficulty —)1.2s

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 1.2s

1 · TTS said no speech captured

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 2026-07-30T20:48:21.402Z

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 margarita." (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 no speech captured

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

3 · App data rows written created food_log_entry: margarita ×1 (serving) 200 cal · 5g P · 20g C · 10g 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 2026-07-30T20:48:32.685Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 sherbet." (difficulty —)0.3s

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 0.3s

1 · TTS said no speech captured

2 · Card shown I logged Sherbet, All Flavors — let me know if that's not right.

3 · App data rows written created food_log_entry: Sherbet, All Flavors ×1 (1 cup) 252 cal · 1.9g P · 53.2g 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 2026-07-30T20:48:44.154Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 seeds." (difficulty —)1.5s

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 1.5s

1 · TTS said no speech captured

2 · Card shown Which one did you mean? 1. Mixed Seeds Brittle — EL SUPER LEON, 2. Mixed Seeds Brittle — El Leoncito G.S. Com, 3. Chewy Muesli Bar, Oats Cranberries & Mixed Seeds — FREEDOM FOODS, 4. Irish Mixed-Seed Crackers Made In Ireland With Unbleached Wheat Flour, Butter, And Mixed Seeds — SHERIDANS CHEESEMONGERS

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 2026-07-30T20:48:56.754Z

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 popcorn." (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 no speech captured

2 · Card shown How much popcorn was it, and was it air-popped, microwave, movie theater, or made with oil or butter? I did not log it yet because popcorn amount and preparation 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 2026-07-30T20:49:08.104Z

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 Aloha bar." (difficulty —)0.5s

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 0.5s

1 · TTS said no speech captured

2 · Card shown Logged RXBAR Mixed Berry Protein Bar — the macros are an estimate since this food varies by brand.

3 · App data rows written created food_log_entry: Mixed Berry Protein Bar ×1 (1 bar) 210 cal · 12g P · 24g 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 2026-07-30T20:49:19.754Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 labneh cup." (difficulty —)0.2s

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 0.2s

1 · TTS said no speech captured

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

3 · App data rows written created food_log_entry: Labneh ×1 (1 cup) 192 cal · 12g P · 6g C · 14.4g 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 2026-07-30T20:49:46.404Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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 nut milk." (difficulty —)5.3s

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 5.3s

1 · TTS said no speech captured

2 · Card shown What fat level or type of milk should I use? I did not log it yet because milk fat level 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 2026-07-30T20:50:02.853Z

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 a burrito." (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 no speech captured

2 · Card shown What exact food and amount should I use for a burrito? 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 2026-07-30T20:50:16.653Z

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 meal shake." (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 no speech captured

2 · Card shown What exact food and amount should I use for a meal shake? 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 2026-07-30T20:50:30.503Z

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 cold brew with cream." (difficulty —)1.7s

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 1.7s

1 · TTS said no speech captured

2 · Card shown I found cold brew, but I also heard cream. 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 2026-07-30T20:50:43.270Z

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 biscuit." (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 no speech captured

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 2026-07-30T20:50:55.420Z

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 turkey burger." (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 no speech captured

2 · Card shown What exact food and amount should I use for a turkey 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 2026-07-30T20:51:09.237Z

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 personal pizza." (difficulty —)5.3s

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 5.3s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for Pizza? 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 2026-07-30T20:51:26.228Z

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 biscuit sandwich." (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 no speech captured

2 · Card shown Was it plain, or did you have anything on it like butter or cream 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 2026-07-30T20:51:39.719Z

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 cold brew size." (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 no speech captured

2 · Card shown What exact food and amount should I use for my usual cold brew size? 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:

Seed food: cold brew

{
  "type": "food_item",
  "foodItem": {
    "id": "seed-f-brew-confirm",
    "name": "cold brew",
    "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 cold brew size",
  "foodItemId": "seed-f-brew-confirm",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "log my usual cold brew size",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:51:53.619Z

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 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 no speech captured

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-f-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-f-dinner-portion",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "add the regular dinner portion",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:52:04.819Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 no speech captured

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-f-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-f-snack-amt",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "track my default afternoon snack",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:52:16.053Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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 no speech captured

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-f-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-f-bfast-size",
  "settings": {
    "quickLogMode": "fixed",
    "defaultQuantity": 1,
    "defaultServingLabel": "serving",
    "displayLabel": "i had the usual breakfast size",
    "confirmBeforeLog": false
  }
}

snapshot captured 2026-07-30T20:52:27.251Z

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

7 · Lookup method The answer used provisional food evidence from user default.

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

PASSshould complete the pending action"Three quarters of a cup." (difficulty —)1.9s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. PASS: Completed the pending action.

Why verdict Completed 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 1.9s

1 · TTS said no speech captured

2 · Card shown I couldn't find Quarter's A Packaged Food in my nutrition database, so I haven't logged it — I'd only be guessing the nutrition. Say the exact name on the package, or 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 2026-07-30T20:52:40.352Z

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)

PASSshould complete the pending action"One and a half cups, plain." (difficulty —)1.3s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. PASS: Completed the pending action.

Why verdict Completed 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 1.3s

1 · TTS said no speech captured

2 · Card shown I couldn't find A Half Half Packaged Food in my nutrition database, so I haven't logged it — I'd only be guessing the nutrition. Say the exact name on the package, or 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 2026-07-30T20:52:52.719Z

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 Aloha chocolate mint." (difficulty —)1.5s

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 1.5s

1 · TTS said no speech captured

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: none listed in the corpus row.

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

snapshot captured 2026-07-30T20:53:05.419Z

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"Unsweetened almond milk." (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 no speech captured

2 · Card shown How much should I log for Unsweetened almond milk? 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 2026-07-30T20:53:16.650Z

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"No, that's not it — the unsweetened one." (difficulty —)5.8s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. 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 5.8s

1 · TTS said no speech captured

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 2026-07-30T20:53:33.635Z

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 complete the pending action"Wrong one. The cherry, not grape." (difficulty —)0.6s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. PASS: Completed the pending action.

Why verdict Completed 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 0.6s

1 · TTS said no speech captured

2 · Card shown Logged Cherries and Grapes. 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: Cherries ×1 (154 g) 97 cal · 1.7g P · 24.6g C · 0.3g F
created food_log_entry: Grapes ×1 (100 g) 69 cal · 0.7g P · 18.1g 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 2026-07-30T20:53:45.485Z

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

7 · Lookup method The answer used provisional food evidence from the common-food list.

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

FAILshould complete the pending action"No, not the flavored labneh." (difficulty —)0.1s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. 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 no speech captured

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 2026-07-30T20:53:56.686Z

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 decline"Cancel that entry." (difficulty —)0.0s

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 0.0s

1 · TTS said no speech captured

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-f-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-f-corr-pending-3 at 2026-07-18T11:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-corr-pending-3",
  "quantity": 1,
  "loggedAt": "2026-07-18T11:00:00.000Z"
}

snapshot captured 2026-07-30T20:54:23.285Z

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 decline"Never mind, remove that." (difficulty —)1.9s

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 1.9s

1 · TTS said no speech captured

2 · Card shown Understood. I’ve cleared the previous request. What would you like to focus on instead?

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-f-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-f-corr-pending-4 at 2026-07-18T11:00:00.000Z

{
  "type": "food_history",
  "foodItemId": "seed-f-corr-pending-4",
  "quantity": 1,
  "loggedAt": "2026-07-18T11:00:00.000Z"
}

snapshot captured 2026-07-30T20:54:36.266Z

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"No, that's not the bar I meant." (difficulty —)0.1s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. 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 no speech captured

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 2026-07-30T20:54:47.452Z

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 the large — the small tangerine." (difficulty —)2.5s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. 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 2.5s

1 · TTS said no speech captured

2 · Card shown I could not finish that request safely.

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 2026-07-30T20:55:01.083Z

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 seared ahi." (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 no speech captured

2 · Card shown What exact food and amount should I use for Ahi tuna? 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 2026-07-30T20:55:17.601Z

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"That's not it; the plain one." (difficulty —)5.8s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. 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 5.8s

1 · TTS said no speech captured

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 2026-07-30T20:55:45.753Z

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"No, the other tortilla." (difficulty —)0.1s

Verdict Expected CLARIFY_ANSWER — should complete the pending action. 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 no speech captured

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 2026-07-30T20:55:56.968Z

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"Not those pistachios — the roasted ones." (difficulty —)5.2s

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 5.2s

1 · TTS said no speech captured

2 · Card shown What exact food and amount should I use for Pistachios? 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 2026-07-30T20:56:24.501Z

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