audit-public-rules

Diagnose AGENTS.md and CLAUDE.md files against best-practice benchmarks.

1|Updated Feb 12, 2026
One-click install
npx skills add https://github.com/dev-goraebap/grimoire --skill audit-public-rules
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit-public-rules
Source: https://github.com/dev-goraebap/grimoire/tree/main/agent-cowork/skills/audit-public-rules
Command: npx skills add https://github.com/dev-goraebap/grimoire --skill audit-public-rules

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Diagnose public guideline files (Track A: root AGENTS.md + Claude bridge; Track B: root CLAUDE.md) against best-practice benchmarks, report issues, and obtain user consent before optional improvements. The track is auto-detected from the existence of root files and the report header shows Detected track: X. It performs anti-pattern checks (e.g., vague advice, linter overlaps, content that reads like code, outdated references, cross-tool duplication, leakage of private settings), token efficiency checks, structure checks, and reference path validation, offering migration options when both tracks exist.

Core Features & Use Cases

  • Auto-detects the active track from root files (AGENTS.md / CLAUDE.md) and validates associated bridges.
  • Performs anti-pattern detection, references validation, token-efficiency checks, and boundary/structure analysis.
  • Suggests track migrations to consolidate rules into a single track when both exist.
  • Produces a guided diagnostic report with actionable recommendations and a user-consent workflow.

Quick Start

Run the audit to generate a diagnostic report for AGENTS.md and CLAUDE.md.

Frequently Asked Questions about audit-public-rules

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit AGENTS.md and CLAUDE.md files against best-practice guidelines?

The audit mechanism auto-detects the active track from root files, then runs anti-pattern checks, token-efficiency analysis, and reference path validation to generate a diagnostic report with actionable recommendations.

What anti-patterns are detected in public guideline markdown files?

Anti-patterns detected in public guideline markdown files include vague advice, linter overlaps, code-like content, outdated references, cross-tool duplication, and leakage of private settings, alongside token-efficiency and structural boundary checks.

How do I migrate between AGENTS.md and CLAUDE.md guideline tracks?

Migration options between AGENTS.md and CLAUDE.md tracks are suggested when both root files exist, allowing you to consolidate rules into a single track through a guided diagnostic report and user-consent workflow before applying changes.

Can I validate reference paths in markdown guidelines automatically?

Yes, you can validate reference paths in markdown guidelines automatically. The diagnostic process includes boundary and reference validation to ensure all linked paths in AGENTS.md and CLAUDE.md files are accurate and not outdated.

Does the guideline audit require manual track selection for AGENTS.md or CLAUDE.md?

No, track selection is fully automatic. The diagnostic detects the active track from the existence of root files (AGENTS.md or CLAUDE.md) and validates associated bridges without requiring manual configuration.