Menu
Use when verifying that a skill actually changes Claude's behaviour — runs TDD pressure scenarios against a skill before and after writing it.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Hostile audit toolkit: one entry point dispatching specialist commands — adversarial review, security, tests, docs, architecture, performance, dependencies, error handling, CI, commits, migrations, observability, API contracts, a11y, privacy, config, resource leaks, i18n, concurrency, unwired code, PR review and review-comment implementation. Use when auditing or reviewing a repository, PR, or any quality dimension of a codebase — "audit this", "review the whole codebase", "find all problems", "exhaustive review", "/nitpicker <command>", a release gate check, or any specific audit ask (security scan, find race conditions, audit the tests, review the PR, fix the CR comments).
Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community detection, and query/path/explain tools.
Routes audit requests to the right /nitpicker command. Use when the user wants to run one of the hostile audit commands in this repo, or asks what audit commands are available.
Use when verifying all skills in the repository are well-formed before a release or after adding/editing a skill.
Use when creating a new hostile audit command for the nitpicker skill in this repository.
Use when validating that all changes in the current branch are ready to be included in a release PR managed by release-please automation.
| name | skill-tester |
| description | Use when verifying that a skill actually changes Claude's behaviour — runs TDD pressure scenarios against a skill before and after writing it. |
| disable-model-invocation | true |
TDD for documentation: watch the agent fail without the skill, write the skill, watch the agent pass.
Dispatch a subagent with this prompt, substituting <skill-name> and <scenario>:
You are working on <scenario>. Do NOT load any skills. <pressure>
What do you do?
Pressure types to combine:
Record exact rationalizations the agent uses to skip the rule.
Write the skill or command body (skills/<skill-name>/SKILL.md, or
skills/nitpicker/commands/<command>.md for a nitpicker command). Address each rationalization from RED explicitly. Then dispatch the same subagent again, this time with the skill loaded. Confirm each RED rationalization is blocked. If a new loophole emerges, add an explicit counter to the skill and re-run.
Refactor the skill body for clarity and precision. Then dispatch the same scenario again (skill still loaded). Confirm all GREEN scenarios still pass and no new loopholes have appeared. If they have, add counters and re-run.
uv run scripts/validate-skill.py skills/<skill-name>/SKILL.md (for a nitpicker command: skills/nitpicker/SKILL.md — it validates the command files too)