--- name: maintain-verification-skill description: "Periodic pass that keeps a project's verification skill and feature map honest: parallel source readers per feature, one live session driving every feature, at most one PR of proven corrections. Use for /maintain-verification-skill or \"audit the verify skill\"." 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. ## Host bindings and permissions Read [host capabilities](../../references/host-capabilities.md), [orchestration contract](../../references/orchestration-contract.md), and [workflow ownership](../../references/workflow-ownership.md). Follow the active owner. Bind reads, independent workers, execution, app driving, writes and external actions to exposed tools and observed results. Documentation alone proves no capability. Discover the current host's project-local skill location from repository conventions and exposed documentation. Do not assume a vendor-specific directory or registration. If the target cannot be found, follow step 0. If independent workers are unavailable, read every feature serially and report reduced capability; serial readers are not independent workers. Keep the same source and live proof standards. Follow the assignment contract for worker identity, revision, ownership, result inspection and retries. Children remain read-only. If execution or app driving is denied or unavailable, retain source findings and unrun recipes, mark live verification `not-run` and report `blocked`. If writes are denied, report recommended corrections without editing. Installation, listeners, credentials, external writes and shipping require applicable authorization. Do not schedule recurring upkeep from this invocation. A changed supplied app contract needs source and live reconciliation; never silently lower the proof requirement to accommodate drift. ## Outcomes Pick one, and say which: - **clean**, every feature got source and live coverage; nothing worth shipping. No branch, no PR. - **changed**, one authorized PR ships 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 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 (in the observed project-local skill directory). Several candidates, 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.** One read-only subagent per feature file, launched concurrently. 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. 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.** Fix wrong or missing user-POV descriptions as doc drift. Fix working behavior the harness can't drive as a harness gap. Harness fixes follow the same helpers rule as generation: scripts are executable and the skill body documents their invocation. Record broken app behavior as a product gap for the user and keep it out of this PR. 6. **Ship or stop.** For changed: re-read every changed file first, then ship one PR of proven corrections only with explicit authorization. Without shipping authorization, retain the proven corrections for local review and report blocked shipping. Do not create a branch or PR. For clean or blocked: no PR, report the outcome and the coverage honestly. Keep concise run notes (features covered, unreachable prerequisites, confirmed drift, outcome) in a scratch location; don't commit them.