---
name: maintain-verification-skill
description: "Audit and repair a project-local verification skill and feature map against source and live behavior. Use when asked to maintain a verification skill or audit an existing verify workflow."
disable-model-invocation: true
---

# Maintain a verification skill

A feature map rots the moment the app changes. This skill is the upkeep loop for a skill generated by `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, without terminalising every bullet.

## Outcomes

Pick one, and say which:

- **clean** — every feature got source and live coverage; nothing worth shipping. No branch, no PR.
- **changed** — proven doc, harness, or map corrections are ready in the working tree; if PR creation is part of the authorized workflow, ship them in at most one PR.
- **blocked** — coverage could not finish or a proven fix could not be completed within scope. Say exactly what blocked it.

## Edit scope

Only edit the verification skill's own directory (its SKILL.md, features/, and any 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.** Find the verification skill to maintain: the project-local skill whose body has launch/drive sections and a feature map (usually `.agents/skills/verify-*/`). Several candidates with no clear target in the request → ask which one; none → stop and point at `create-verification-skill` instead of inventing a target.

1. **Index hygiene.** Read the feature map README and glob its sibling files. Fix missing, extra, duplicate, or dead entries. Lightweight; no generated inventory.

2. **Source wave.** Use read-only subagents for independent feature files within the host’s available concurrency; batch them when capacity is limited. If delegation is unavailable, perform the same source pass sequentially. Each explains "how does this user-facing feature work?" from source, flags likely doc drift with 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 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. Use the host’s available tools and normal permission controls; an unavailable required capability is a coverage blocker. The coordinator owns all driving; follow the verification 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 (the skill's Launch section decides, not this one). Exercise every feature at least once, and hold three invariants the whole pass, whatever the failure: (1) never drive an instance you haven't health-checked since it last did something surprising — doctor before first drive, doctor on each fresh session where sessions are the unit, doctor again after any failed drive, and where doctor can't see the failure (a wedged UI state on a healthy process), reset to a known state or relaunch rather than hoping; (2) evidence captured so far survives every cleanup, checked at its named location, not assumed; (3) nothing a drive started outlives that drive's usefulness — failed-iteration residue is cleaned whether the session is stuck, exited, or shared (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 — restart whatever the fix invalidated, nothing more — before calling the pass `blocked`. A feature that can't be reached 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. Any harness fix from triage gets re-driven live before it ships. Final teardown happens after the last drive of the run — including those re-proofs — so nothing outlives the run (evidence stays, per the skill).

5. **Triage.** Wrong or missing user-POV description → doc drift, fix it. Working behavior the harness can't drive → harness gap, fix it; a harness fix follows the same helpers rule as generation (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 these verification corrections.

6. **Finish or ship.** Re-read every changed file and report the proven corrections with coverage and evidence. Create at most one PR only when requested or already authorized by the task’s workflow; otherwise leave the changes locally for review. For clean or blocked: no PR, report the outcome and coverage honestly. Do not schedule recurring maintenance unless requested.

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