gate-risk-reviewerlisted

# Gate Risk Reviewer Skill ## Intent Promotion must be based on evidence, policy, and REQ satisfaction. This skill prevents generated prose, optimistic assumptions, partial success, or `PASS_WITH_WARNINGS` from being treated as promotable output. ## Use when Use this skill for every GATE phase, every EVAL summary review, every FINALIZE readiness review, and any REQ touching runtime safety, public contracts, persistence, integrations, security, UI behavior, AI behavior, industrial workflows, mobile offline behavior, or deployment. ## Do not use when Do not use this skill as a subjective style reviewer. It must not block promotion for taste, formatting preference, or architectural opinion unless the issue violates a policy, acceptance criterion, selected capability, or executable evidence requirement. ## Signals Apply this skill when any of these are present: - generated source code; - generated tests; - generated LTC/HOWTO; - public API changes; - auth/security changes; - persistence changes; - external integrations; - frontend workflows; - AI/RAG/tool behavior; - runtime profile requirements; - selected skills/packs/design profiles; - warnings from EVAL; - missing reports; - missing commands; - partial evidence. ## Gate Decision Model Gate must evaluate in this order: 1. REQ dependency status. 2. Functional acceptance criteria. 3. Technical acceptance criteria. 4. Required local validation evidence. 5. Runtime profile adherence. 6. Selected skill adherence. 7. S