# Audit

> Lint the ADRs and judge the code in one run, over a diff or the whole codebase. Read-only.

- Skill: `rvdbreemen/audit-2` (Agent Skill)
- Install (CLI): `npx skillmds@latest add rvdbreemen/audit-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/rvdbreemen/audit-2/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- License: MIT
- Author: rvdbreemen (https://skillmd.com/u/rvdbreemen)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/rvdbreemen/audit-2

---


<!-- Generated by scripts/build-client-adapters.py from clients/workflows.json; schema v1; client codex-cli. -->
# Audit ADRs and code

Invoke explicitly with `$adr-kit:audit` when needed.
Resolve `<plugin-root>` from this installed skill. Use only bundled,
local ADR Kit tools and follow this canonical workflow:

1. Run `python <plugin-root>/bin/adr-audit --whole-codebase --adr-dir docs/adr --repo-root .`, or `--diff -` with a piped unified diff for the hook and pull-request case.
2. Read the exit code before the output: 0 clean, 1 the code violates an Accepted ADR, 3 the ADR set fails its own gates, 4 both, 2 the audit could not run.
3. Report the two failures separately and name their owners; never collapse a code violation and an ADR-quality failure into one verdict, and never present exit 2 as a pass.
4. Add `--llm` only with explicit user confirmation, because it spends money per opted-in ADR.

Do not contact another model or mutate user-owned instructions.