# Guardian

> Run ADR Kit health, drift, stale-decision, and optional semantic sweeps for the active project.

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

---


<!-- Generated by scripts/build-client-adapters.py from clients/workflows.json; schema v1; client codex-cli. -->
# ADR guardian

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

1. Run `adr-doctor`, `adr-retire`, `adr-status`, and `adr-guardian retrieval-health` against the active workspace and report drift or failed retrieval probes first.
2. Rank Proposed ADRs by active implementation link, shipped-but-Proposed, ready-for-confirmation, open human questions, age, and lowest quality; show at most three client-native grill invocations.
3. Use local-model judgment automatically when configured; require explicit opt-in before a paid or cloud model.

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