AU
audit
Audit a new or unfamiliar project to understand its structure, health, risks, documentation gaps, and next actions. Use when initially entering a repo, assessing project quality, preparing onboarding, or asking what should be improved first.
Install
mkdir -p .claude/skills/audit-hunvreus && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/16382" && unzip -o skill.zip -d .claude/skills/audit-hunvreus && rm skill.zipInstalls to .claude/skills/audit-hunvreus
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Audit a new or unfamiliar project to understand its structure, health, risks, documentation gaps, and next actions. Use when initially entering a repo, assessing project quality, preparing onboarding, or asking what should be improved first.241 chars✓ has a “when” trigger
About this skill
Audit
Input
- A repository, path, subsystem, branch, feature, diff, or audit question.
- Use explicit input first; otherwise infer from context, recent edits, selected files, or branch.
- Safest default: audit the current project at a high level.
Workflow
- Map the project. Identify purpose, stack, entry points, package structure, runtime, and ownership boundaries.
- Find commands. Locate install, dev, test, lint, typecheck, build, release, and deploy commands.
- Run safe checks. Execute available non-destructive checks when practical; record missing or failing checks.
- Review docs. Check README, architecture docs, API docs, devops docs, runbooks, and examples for gaps or stale claims.
- Inspect risk areas. Review security, dependencies, config, migrations, data handling, error paths, observability, and release/deploy posture.
- Identify issues. Produce prioritized findings with evidence, impact, and suggested next action.
- Fill obvious docs gaps. If requested, update small missing docs discovered during the audit.
Output
- Project map
- Available commands and check results
- Findings ordered by severity or leverage
- Documentation gaps
- Suggested next actions
- Work explicitly not audited
Guardrails
- Do not make broad changes during an audit unless the user asks.
- Do not run destructive commands or deploys.
- Distinguish observed facts from guesses.
- Prefer actionable findings over generic best-practice lists.