agentleFS
GTStaggGTStagg/project-eyas/.claude/skills/verify-patch/SKILL.md

verify-patch

Skill2 starsChanged yesterdayMITView on GitHub

Use after writing or editing a patching script + manifest entry (lib/eyas-tools/rtr/patching). Runs the canonical render + spotlight-match + non-ASCII smoke test for one or more script ids. Trigger phrases: "verify the script", "smoke-test the patch", "does the new manifest entry render", "check spotlight matching".

The file

Files
SKILL.md

verify-patch

Smoke-test one or more patching catalog entries end-to-end before considering the work done.

Inputs

The user (or you, post-edit) supplies a script id (e.g. app-zoom) and optionally a list of (vendor, product) pairs the entry should resolve to. If no spotlight cases are given, infer two or three plausible Spotlight remediation strings from the script's purpose and check those.

What to run

From the repo root, invoke the verifier:

bash .claude/skills/verify-patch/run.sh <script_id> [<script_id> ...]

The script (scripts/rtr-library-cli.mjs verify) renders each plan via resolvePlan in lib/eyas-tools/rtr/library.ts, prints body length / runtime / reboot risk / non-ASCII byte count, and confirms the spotlight pattern matches the demo cases bundled in the script.

Pass criteria

  • resolvePlan returns without exception
  • non-ascii=0 on every rendered body (RTR strips non-ASCII bytes — they break the host parser)
  • For entries with a spotlight_match, the canonical demo (vendor, product) pair routes back to the same id (regression-tests first-declared-wins ordering)

Fail handling

  • unknown script id → manifest entry missing or typo'd; re-check manifest.json.
  • non-ascii > 0 → asciiCoerce is the backstop, but source files should be clean. grep -nP '[^\x00-\x7F]' <file> to find offenders.
  • Wrong id resolved by matchRemediation → another manifest entry's pattern is shadowing yours. Move the new entry earlier in the manifest, or tighten the colliding pattern.
  • A TypeScript error from the CLI → the library API drifted; re-read lib/eyas-tools/rtr/library.ts.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.