# Maintain Verification Skill

> Periodic pass that keeps a project's verify-<app> skill and feature map honest: parallel read-only source readers per feature, one live session driving every feature, at most one change set of proven corrections. Use for /skill:maintain-verification-skill or "audit the verify skill".

- Skill: `timothyvang/maintain-verification-skill` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx get-skill add timothyvang/maintain-verification-skill`
- Raw SKILL.md: https://api.skillmd.com/api/skills/timothyvang/maintain-verification-skill/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: timothyvang (https://skillmd.com/u/timothyvang)
- Updated: 2026-11-16
- Page: https://skillmd.com/skills/timothyvang/maintain-verification-skill

---


# Maintain a verification skill

A feature map rots the moment the app changes. This is the upkeep loop for a skill generated by `/skill:create-verification-skill` (or any project-local verification skill with a feature map). The unit of rigor is the feature, not every sentence: cover every feature file from source and exercise every feature live.

## Outcomes

Pick one and say which:

- **clean**: every feature got source and live coverage; nothing worth changing.
- **changed**: one change set of proven doc, harness, or map corrections.
- **blocked**: coverage could not finish, or a proven fix could not ship safely. Say exactly what blocked it.

## Edit scope

Only edit the verification skill's own directory (its `SKILL.md`, `features/`, and harness scripts it owns). Never edit product code during a run. A behavior the map describes that the app no longer does is either doc drift (fix the map) or a product regression (report it; don't paper over it in docs).

## Pass

0. **Locate the target.** Glob `.omp/skills/verify-*/SKILL.md`, `.claude/skills/verify-*/SKILL.md`, and `.agents/skills/verify-*/SKILL.md` from the repo root (omp, Claude Code, and Codex and most other harnesses read those folders). Several candidates → pick the one whose app the change touches, and log why. None → stop and point at the **create-verification-skill** skill instead of inventing a target.

1. **Index hygiene.** Read `features/README.md` and glob its sibling files. Fix missing, extra, duplicate, or dead entries.

2. **Source wave.** One read-only `scout` task per feature file, in a single `task` batch. Each explains "how does this user-facing feature work?" from source, flags likely doc drift with file:line citations, and returns one concise live-verification recipe. Children never drive the app and never edit files. Return shape: feature summary / source entry points / likely drift or none / one recipe.

3. **Reconcile.** Every feature file has a returned summary. Merge overlapping recipes into as few app states as practical. Spot-check cited drift; don't re-prove clean claims. Sweep recent churn (`git log --since`) for user-facing surfaces missing from the map; require a concrete source path before calling one missing.

4. **Live pass.** Required even when source looks clean. You (the coordinator) own all driving. Follow the skill's own launch model: one long-lived instance driven serially for servers and UIs, or a fresh isolated session per drive for short-lived CLIs. Exercise every feature at least once. Hold three invariants whatever fails:
   - Never drive an instance you haven't doctored since it last did something surprising: doctor before the first drive, on each fresh session, and after any failed drive. Where doctor can't see the failure (a wedged UI on a healthy process), reset or relaunch rather than hoping.
   - Evidence captured so far survives every cleanup, checked at its named location, not assumed.
   - Nothing a drive started outlives its usefulness. For a shared instance, clean the residue, not the instance.

   A doctor failure caused by skill drift is drift: fix it under edit scope and retry once, restarting only what the fix invalidated, before calling the pass `blocked`. A feature is `verified-unreachable` only with the concrete prerequisite (auth, entitlement, OS, external state) and the route attempted; if the map omits that prerequisite, that's drift. Re-drive every harness fix live before it ships. Final teardown happens after the last drive, re-proofs included; evidence stays.

5. **Triage.**
   - Wrong or missing user-POV description → doc drift; fix it.
   - Working behavior the harness can't drive → harness gap; fix it (scripts executable, invocation documented in the skill body).
   - App behavior that's actually broken → product gap; record it for the user, keep it out of this change.

6. **Ship or stop.** For **changed**: re-read every changed file, then present one change set. Commit it and open a PR per deepmodebot's Opening a PR playbook (`skill://deepmodebot/references/playbooks/opening-a-pr.md`). For **clean** or **blocked**: no changes; report the outcome and coverage honestly.

Keep run notes (features covered, unreachable prerequisites, confirmed drift, outcome) in a scratch location; don't commit them.

