| name | pmstack-launch-readiness |
| description | Pre-launch verifier. Aggregates PRD, metrics, eval design, eval run, premortem, eval-drift, and brief evidence into a single GO/NO-GO/CONDITIONAL verdict with each item showing pass/fail/missing plus the file that proves it. Trigger when the user says "are we ready to ship," "launch readiness," "release check," "pre-launch review," "ship it?" or asks to gate a release. Run before any launch. |
launch-readiness — pre-launch evidence gate
A verifier that reads the artifact graph for a feature and returns GO / NO-GO / CONDITIONAL.
Required structure
# launch-readiness — <feature> — <YYYY-MM-DD>
VERDICT: GO|NO-GO|CONDITIONAL
## Checklist
| Item | Status | Evidence |
|---|---|---|
| PRD signed off | Pass/Fail/Missing | <path or "—"> |
| Metrics defined | Pass/Fail/Missing | <path or "—"> |
| Eval designed | Pass/Fail/Missing | <path or "—"> |
| Eval actually run | Pass/Fail/Missing | <path or "—"> |
| No P0 regression (latest eval-drift) | Pass/Fail/Missing | <path or "—"> |
| Premortem completed | Pass/Fail/Missing | <path or "—"> |
| Brief sent | Pass/Fail/Missing | <path or "—"> |
## Verdict reasoning
<2–3 sentences naming the failures or gaps>
Verdict rules
- GO — all 7 items Pass.
- NO-GO — any P0 item is Fail (PRD missing, eval-drift RELEASE_BLOCKED).
- CONDITIONAL — any non-P0 items are Missing or Fail.
Hard rules
Where to write
- With filesystem:
outputs/launch-readiness-<feature>-<YYYY-MM-DD>.md.
- Inline (web/mobile): emit the artifact as markdown with the suggested filename at the top.
Decision-log entry
Always: - <date> — launch-readiness(<feature>): <VERDICT> — <path>. On override, an additional SHIPPED-DESPITE-... line.
Tone
Honest, three-state, not punitive. The verdict is information, not a judgment.
Cold-start behavior
Empty outputs/ correctly returns NO-GO with all items Missing. That's the routine working as designed.
