audit-rules
Audit Claude Code rule files for quality and compliance. Use when creating or validating .claude/rules/*.md files, or troubleshooting rule loading issues.
Best use case
audit-rules is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Audit Claude Code rule files for quality and compliance. Use when creating or validating .claude/rules/*.md files, or troubleshooting rule loading issues.
Teams using audit-rules should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/audit-rules/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How audit-rules Compares
| Feature / Agent | audit-rules | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Audit Claude Code rule files for quality and compliance. Use when creating or validating .claude/rules/*.md files, or troubleshooting rule loading issues.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Audit Rules Command Audit Claude Code rule files (`.claude/rules/*.md`) for quality and compliance. ## Initialization Before auditing, initialize the environment: 1. Get the current UTC date for audit timestamps. 2. Capture the project root path for subagent communication. 3. Ensure the temp directory (`.claude/temp/`) exists. 4. Clean up any stale audit files if the user confirms. The `memory-management` skill provides authoritative validation guidance for rules (auto-loaded when this command runs). ## What Gets Audited - YAML frontmatter structure (description, globs) - Glob pattern validity and syntax - Rule file naming conventions - Content structure and clarity - Path-specific rule applicability ## Command Arguments | Argument | Description | | --- | --- | | *(none)* | Audit all discoverable rule files | | `--force` | Audit regardless of modification status | | `--skip-validation` | Skip finding validation (faster, but may include false positives) | ## Step 1: Discover Rule Files Search for rule files in: - Project rules (`.claude/rules/*.md`) - User rules (`~/.claude/rules/*.md`) Build a list of discovered rule files with their scope (project or user) and full path. If no rule files are found, report this and provide guidance on how to create one. ## Step 2: Parse Arguments Check if the `--force` flag is present in the command arguments. Build the audit queue based on discovered files and the force flag. ## Step 3: Present Audit Plan Display audit mode (SMART or FORCE), rule files discovered, and list each with scope and last modified date. ## Step 4: Execute Audits For each rule file, spawn the `memory-component-auditor` subagent with the following context: - Scope (project or user) - Full path to the rule file - Last audit date or "Never audited" - Current audit date - Project root path Run subagents in parallel when multiple rule files exist. Subagents write findings to `.claude/temp/` as both JSON (for recovery/aggregation) and markdown (for human review). The main conversation thread collects results and updates audit logs using its Write/Edit tools. ## Step 4.5: Validate Findings **Unless `--skip-validation` flag is present:** 1. Spawn the `audit-finding-validator` agent with: - `project_root`: The captured project root path - `audit_type`: "rule" - `audit_files`: List of `.claude/temp/audit-*-rule-*.json` file paths 2. Wait for validation to complete 3. Read updated JSON files with validation results 4. Filter out FALSE_POSITIVE findings completely before aggregation 5. Note: Filtered findings are logged to `.claude/temp/audit-filtered-findings.json` **If `--skip-validation` flag is present:** - Skip validation phase entirely (current speed preserved) - Present all findings without filtering - Note in summary: "Validation: Skipped" ## Step 5: Final Summary Report total rule files audited, results by scope, and details table. List frontmatter or glob pattern issues with remediation steps. **Include validation statistics (if validation was performed):** - Validation performed: Yes/No - Findings validated: X - False positives filtered: Y - Verified findings: Z - Unverified findings: W ## Important Notes ### Rule File Requirements Rule files must have valid YAML frontmatter with `description` and optionally `globs` fields. The `globs` field controls which files the rule applies to. ### Rule File Locations | Location | Purpose | | --- | --- | | `.claude/rules/*.md` | Project-specific rules | | `~/.claude/rules/*.md` | User-wide rules | ### Glob Pattern Syntax Rules can use glob patterns to apply only to specific files: ```yaml --- description: TypeScript coding standards globs: ["**/*.ts", "**/*.tsx"] --- ``` ## Audit Log Location All audit results are written to `.claude/audit/rules.md`. Use `/audit-log rules` to view current audit status. ## Example Usage ### Example 1: Audit All Rule Files ```text User: /audit-rules Claude: Discovering rule files... ## Audit Plan **Mode**: SMART **Rule files discovered**: 3 1. [project] .claude/rules/typescript.md 2. [project] .claude/rules/security.md 3. [user] ~/.claude/rules/personal-style.md [Spawns memory-component-auditor subagents] ## Audit Complete | Scope | Rule File | Result | Score | | --- | --- | --- | --- | | project | typescript.md | PASS | 100/100 | | project | security.md | PASS | 95/100 | | user | personal-style.md | PASS WITH WARNINGS | 82/100 | ``` ### Example 2: Force Audit ```text User: /audit-rules --force Claude: Auditing all rule files (force mode)... ```
Related Skills
divek-bi-visual-audit
Visual compliance auditing for DiveK brand identity. Use when reviewing UI screens, component libraries, landing pages, design handoff specs, CSS tokens, or visual QA reports for alignment with DiveK color palette, typography, and cinematic-minimal style direction.
dependency-management-deps-audit
You are a dependency security expert specializing in vulnerability scanning, license compliance, and supply chain security. Analyze project dependencies for known vulnerabilities, licensing issues,...
Cookbook Audit
Audit an Anthropic Cookbook notebook based on a rubric. Use whenever a notebook review or audit is requested.
Configure Firebase admin-only write rules
Guide to create an admin user in Firebase and set security rules so only admins can write to a collection while all users can read and register.
Compliance Audit
Audit technical controls against compliance framework requirements
codebase-cleanup-deps-audit
You are a dependency security expert specializing in vulnerability scanning, license compliance, and supply chain security. Analyze project dependencies for known vulnerabilities, licensing issues,...
codebase-audit
Performs comprehensive codebase audit checking architecture, tech debt, security, test coverage, documentation, dependencies, and maintainability. Use when auditing a project, assessing codebase health, or asked to audit/analyze the entire codebase.
claude-a11y-audit
Use when reviewing UI diffs, accessibility audits, or flaky UI tests to catch a11y regressions, semantic issues, keyboard/focus problems, and to recommend minimal fixes plus role-based test selectors.
aws-security-audit
Comprehensive AWS security posture assessment using AWS CLI and security best practices
auditor-workflow
Group-level implementation audit workflow for auditor agents. Handles loading project rules, reading connected phases, reviewing code reviews, checking deferred items, cross-phase impact analysis, verification, and structured reporting to the orchestrator. Invoke this skill as your first action — not user-invocable.
auditor-gate
Apply final governance and release-gate checks to a judged change set by reading `handoff.json`, `verdict.json`, optional eval evidence, and emitting machine-readable `audit.json` with `gate` status. Use when implementation already has a judge verdict and a separate auditor must decide landability (`pass`, `fail`, or `needs-human`) without modifying source files.
auditing-accessibility-wcag
Checks components and pages for WCAG 2.1 accessibility violations. Use when the user asks about a11y, WCAG compliance, screen readers, aria labels, keyboard navigation, or accessible patterns.