gh0stinthemirr0r/gaeos-creator/vet-changes
- community
Checking that the SKILL.md file is still in the repository.
Verify a code change or agent action before accepting it. Checks intent (does it do what was asked), correctness (does it run / type-check / pass tests), scope (no unintended edits, no stray files, no secrets), and regressions (did anything that worked before break). Use after any non-trivial edit, before committing, or when reviewing agent-produced work. Do NOT use for style-only linting or for security penetration testing.
Install
npx vskill@latest install gh0stinthemirr0r/gaeos-creator/vet-changesSKILL.md
Vet Changes — Verify Everything
Never accept a change on faith. Before declaring work done, run the four checks below and only report success when each one passes. Record what you checked and what you found.
1. Intent — does it do what was asked?
- Restate the original request in one sentence.
- Re-read the diff and confirm every hunk serves that request.
- Flag anything the request did not ask for: refactors, renames, dependency bumps, formatting sweeps, "while I was here" fixes.
- If the change solves a different problem than the one posed, it fails intent even if it is technically correct.
2. Correctness — does it actually work?
- Run the project's real checks; never assume. Discover them first (README,
package.jsonscripts,Makefile,pyproject.toml, CI config). - At minimum: compile/type-check, lint, and the relevant test suite.
- A change with no runnable verification is not verified — say so explicitly rather than implying it was tested.
- New behavior needs a new test or an executed manual check; record the command and its output.
3. Scope — is the footprint what you expect?
- Inspect the full list of touched files; confirm each was intended.
- Ensure no secrets, keys, or credentials were added or logged.
- Ensure no generated artifacts, lockfile churn, or unrelated files were committed.
- Confirm no debug scaffolding, commented-out blocks, or TODO debris was left behind.
4. Regressions — did anything that worked before break?
- Run the broader test suite, not just the tests near the change.
- If a pre-existing failure exists, note it as pre-existing and do not claim it as a new break — but also do not add to it.
- Check the change against its callers/consumers: signatures, return shapes, and assumptions still hold.
Output
Report a short verdict per check (pass / fail / not-run) with the evidence (command + result). If any check fails or was not run, say so plainly. Do not suppress a failure to make the change look complete — fix the change instead.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingengine/skills/vet-changes/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/gh0stinthemirr0r/gaeos-creator/vet-changes)