| name | review |
| description | Run a convergence-check review of the current run without shipping — reviewer reads every artifact across code/test/simulation pipelines and issues a PASS / PASS WITH NOTE / STOP verdict. Use this skill when the user says "review this", "check the work", "audit the run", "give me a verdict", "is it ready to ship", or invokes $review. Runs Workflow 8 (Review Only) — reviewer only, no shipper dispatch. |
| metadata | {"short-description":"Convergence check & ship-or-not verdict — workflow 8"} |
Skill: Review — Convergence check without shipping
You are the StatsClaw leader. The user has triggered $review (or said "review this" / "is it ready" / "audit the run").
Run Workflow 8 (Review Only) from skills/statsclaw-protocol/SKILL.md:
leader → reviewer
Steps
-
Locate the run directory. Use the most recent run under ${STATSCLAW_CODEX_DATA}/workspace/<repo>/runs/ (or the run the user explicitly names).
-
Dispatch reviewer. scripts/dispatch.sh reviewer <run-dir>.
-
Reviewer reads ALL artifacts:
request.md, impact.md, comprehension.md
spec.md, test-spec.md, sim-spec.md (if present)
implementation.md, simulation.md (if present)
audit.md, ARCHITECTURE.md, log-entry.md, docs.md
brain-contributions.md (if present)
And writes review.md with the verdict.
-
Report the verdict (PASS / PASS WITH NOTE / STOP) and routing (which teammate reviewer identified as responsible for any issues) back to the user.
Important
Do NOT dispatch shipper from this skill. This workflow is review-only — the user decides whether to invoke $ship-it afterward.