What problem does it solve? A user-research prior-art survey artifact (a research vocabulary map or a per-angle search output) can look plausible while silently misrecording what was searched and found. This Skill acts as an acceptance gate: it judges a finished artifact against a single shared conditions file before the artifact is accepted, so dishonest zeros, conflated failures, and ungrounded claims are caught instead of passing review. ## Core Features & Use Cases - Condition-based review: Walks the judgment conditions in references/conditions.md in order, grounding every finding in quoted artifact text with a named condition number. - Deterministic gate integration: Runs the producer's validator (validate_user_research_prior_art.py) first so gated conditions are discharged mechanically and not duplicated as findings. - Strict verdict output: Emits exactly one line — VERDICT: approve or VERDICT: revise — followed by all findings in a single pass. - Use Case: After authoring a search output for one research angle, run this reviewer before accepting it; it checks coherence of zero/failure/refusal cells, admission conjuncts (retrievable full text plus a stated method), and coverage accounting, then returns a verdict with actionable findings. ## Quick Start Use reviewing-user-research-prior-art-survey to judge the search output in artifacts/angle-2-search.json against its vocabulary map before I accept it.
