Menu
Use when creating a new hostile audit command for the nitpicker skill in this repository.
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 validating that all changes in the current branch are ready to be included in a release PR managed by release-please automation.
Use when verifying that a skill actually changes Claude's behaviour — runs TDD pressure scenarios against a skill before and after writing it.
| name | new-command |
| description | Use when creating a new hostile audit command for the nitpicker skill in this repository. |
| disable-model-invocation | true |
Creates a public command under skills/nitpicker/commands/<name>.md and
stress-tests it through the full RED → GREEN → REFACTOR → adversarial-review
→ validate cycle before declaring it done.
Before writing a single line of command content, run /skill-tester to
establish the baseline. Use its checklist to have a subagent run without
the new command loaded and record every rationalization the agent uses to
skip the rule you are encoding. Document those rationalizations — every one
must be countered in the command body.
Choose a short kebab-case command name (2.0 vocabulary: perf, deps,
errors — not -auditor/-hunter suffixes).
Create the file skills/nitpicker/commands/<name>.md:
# /nitpicker <name> — <Short Title>
<one-line purpose>
## When to use
<trigger conditions and phrasings>
Required sections after that (no YAML frontmatter — only the router
SKILL.md has frontmatter): the hostile mindset/scope, a deterministic
checklist, domain-specific rules, and domain-specific common mistakes.
Do NOT restate what commands/_conventions.md already binds (severity
table, findings store protocol, generic rules) — findings are filed via
findings.py --auditor <name>.
Counter every RED-phase rationalization explicitly in the command body. Use "rationalization" (US spelling) consistently.
Run /skill-tester again with the command loaded (/nitpicker <name>).
Confirm the agent complies with every rule. If the agent finds a new
loophole, add an explicit counter and re-run until no loopholes remain.
_conventions.md (only if it truly binds every command)./skill-tester after refactoring — all GREEN scenarios must pass.Run /nitpicker review against the new command file. Treat the command body
as code under review — hunt ambiguous rules, missing edge-case coverage,
hedged language, instructions that permit rationalization. Fix every HIGH or
CRITICAL finding and re-run until none remain.
## Commands table in
skills/nitpicker/SKILL.md (the validator enforces table ↔ file 1:1)..claude/skills/skills/SKILL.md (Routing
Guide) and the command row in README.md. Update
.github/copilot-instructions.md only if the command changes a rule
stated there. uv run scripts/validate-skill.py skills/nitpicker/SKILL.md
/validate-skills to confirm everything remains consistent.Run /nitpicker pr on the staged diff. Fix all findings at HIGH or above,
re-running until clean. Commit with feat: add /nitpicker <name> command —
this triggers a minor version bump.
A command is not done until all of these pass:
/nitpicker review: no HIGH or CRITICAL findings remainuv run scripts/validate-skill.py skills/nitpicker/SKILL.md exits 0/validate-skills exits clean/nitpicker pr: no HIGH or CRITICAL findings in the diff