← ClaudeAtlas

meta-reviewlisted

Audits any SKILL.md against SKILL-SPEC.md, the repo's skill authoring standard, and returns a scored checklist with prioritized fixes. Trigger on: "review this skill", "audit this SKILL.md", "does this skill meet spec", "is this skill done", "quality check this skill", or any request to validate a skill file against the repo's authoring standard.
stefanoskarakasis/Product-Marketing-Skills · ★ 13 · AI & Automation · score 86
Install: claude install-skill stefanoskarakasis/Product-Marketing-Skills
# meta-review Audits a skill's `SKILL.md` against `SKILL-SPEC.md` — the one standard every skill in this repo is supposed to meet. Runs the same 15-point checklist `SKILL-SPEC.md` Section 13 defines, returns a pass/fail per check, and prioritizes fixes so the skill's author knows exactly what to do next. This is a spec-compliance auditor, not a content-quality judge. It checks structure — is every required section present, does frontmatter have every required field, is the file under 500 lines — not whether the skill's actual domain logic is good. Judging domain logic requires domain expertise this skill doesn't have; a human or a domain-specific reviewer does that. ## Trigger - **When:** A skill's `SKILL.md` needs to be checked against the repo's authoring standard — before merging a new skill, after editing an existing one, or as a periodic health check. - **Not for:** Checking the correctness of a skill's actual output (a GTM brief, a positioning statement, a retro summary) → use `meta-verify`. Detecting cross-skill patterns or proposing guardrails → use `meta-synthesis`. Extracting learnings from a single completed session → use `meta-learn`. - **Example prompts:** - "Review the retro skill" - "Audit pmm-okrs against spec" - "Does stakeholder-maps meet the authoring standard?" - "Check every skill in pmm-execution" ## Inputs - **Args:** Skill name (required) — the directory name of the skill to audit, e.g. `retro` or `positioning-messaging`.