quality-review

name	quality-review
description	Deep review of any work-product — code, docs, specs, plans, decisions — grounded in current authoritative sources. Use when double-checking against latest docs, verifying versions or claims, checking security, or pressure-testing correctness and elegance before something ships. Complements the automatic quality hook with ecosystem verification. NOT for divergent ideation (brainstorm), weighing still-open options (figure-it-out), your own spec's framing (self-review), or scenario review (review-spec).
allowed-tools	*

Quality Reviewing

Deep review with research to verify a work-product — code, docs, specs, plans, decisions — against current, authoritative sources.

Stakes set depth. Review as if your verdict is the last gate before this ships — no one re-checks behind you. That standard, not "the hook already looked," sets how hard you research. Before searching, write your review plan: which angles (§2–3) this work-product actually needs and the specific question each must answer, then work the list — don't stop at the first finding.

When to use (vs. the automatic hook): the hook does a fast, code-only check from existing knowledge after an edit; this skill adds research (~2-3 min) and works on any work-product. Reach for it on explicit verification ("double check against latest docs", "verify versions", "check security"), deep dives (performance, architecture, trade-offs, an argument's soundness), or pre-change review.

Invocation log

Required before marking done a ticket with two or more RGR loops. The line below logs a current-run entry to skill-invocations.log under the project namespace root so the done-gate hook can verify /quality-review actually ran; Claude Code expands the ! line automatically. On Cursor and Codex the pre-shell hook (beforeShellExecution / PreToolUse) bridges the session id, so the fallback runs on all three runtimes without hand-picking one. Hand-writing review notes cannot produce this gate proof.