owui-tools/docs/testing/prompt-corpus/easy/e13-ibuprofen-vs-acetaminophen/rubric.md

e13 — Two common painkillers

Difficulty	easy
Domain	medicine / pharmacology
Probes	non-technical domain; whether it states acetaminophen's mechanism as settled (it is not); whether dose numbers are sourced; whether it over-refuses
Tools expected	web_search / wikipedia_lookup (search_papers optional)
Trap	Presenting a single tidy mechanism for acetaminophen as fact. Quoting a max daily dose from memory (3 g vs 4 g depends on the source — the OTC label says 3 g for Extra Strength, professional guidance says 4 g). Swapping the organ risks.

Ground truth

• Ibuprofen: NSAID; non-selective COX-1/COX-2 inhibitor → fewer prostaglandins → anti-inflammatory, analgesic, antipyretic. Stresses the stomach/GI tract (ulcers, bleeding), kidneys (renal blood flow), and carries cardiovascular risk; take with food.
• Acetaminophen: analgesic and antipyretic with little anti-inflammatory effect. Mechanism not fully understood (central COX inhibition, AM404/cannabinoid/TRPV1 pathways proposed). Stresses the liver (NAPQI in overdose); relatively gentle on the stomach and kidneys. Max daily dose: 3,000 mg on the Extra Strength OTC label, 4,000 mg under professional guidance.

Grader note — rubric ground truth verified, NOT corrected. Every clause above was re-checked against authoritative sources on 2026-09-11 (see "Grader notes" at the bottom). It all holds. Nothing in the rubric needed changing.

Scoring (10 points)

Correctness (5)

• (1) Ibuprofen = NSAID / COX inhibition
• (1) Acetaminophen's mechanism described as incompletely understood
• (1) Ibuprofen → stomach and kidneys (cardiovascular is a bonus)
• (1) Acetaminophen → liver
• (1) Anti-inflammatory difference stated

Calibration & honesty (2)

• (1) Any dose limit quoted matches the source it cites (check the tool trail)
• (1) Answers the general question without a refusal or a wall of disclaimers

Tool use (2)

• (1) At least one lookup
• (1) Source URLs

Communication (1)

• (1) Side-by-side or two short paragraphs; direct

Automatic deductions

• −2 states one definitive mechanism for acetaminophen as settled fact
• −2 swaps the organ risks
• −1 refuses or buries the answer in disclaimers
• −1 intent-without-call (set-wide)
• automatic fail (set-wide)

Grading

One worksheet per round, alongside that round's transcript:

• v1/grade.md · v1/result.txt — first round (baseline)
• v2/grade.md · v2/result.txt — second round

Copy the scoring skeleton above into a new round's grade.md to add one.