Rules Evaluation Framework
When NOT To Use
- Evaluating skills (use
abstract:skills-eval)
- Evaluating hooks (use
abstract:hooks-eval)
- Writing a new rule (use
hookify:writing-rules)
Table of Contents
- Overview
- Quick Start
- Evaluation Workflow
- Scoring
- Resources
Overview
This skill evaluates Claude Code rules in .claude/rules/ directories against quality standards. It validates YAML frontmatter, glob pattern syntax, content quality, and directory organization. Rules files support path-scoped conditional loading via paths frontmatter and unconditional rules (no paths field).
Key validations: YAML syntax errors, unquoted glob patterns, Cursor-specific fields (alwaysApply, globs), overly broad patterns, content verbosity, and naming conventions.
Quick Start
# Evaluate rules in current project
/rules-eval
# Evaluate specific directory
/rules-eval .claude/rules/
# Detailed analysis with recommendations
/rules-eval --detailed
Evaluation Workflow
- Scan
.claude/rules/ for all .md files (including subdirectories)
- Validate YAML frontmatter syntax and fields
- Analyze glob patterns for correctness and specificity
- Assess content quality (actionable, concise, non-conflicting)
- Check organization (naming, structure, symlinks)
- Measure token efficiency and redundancy
Scoring
| Category |
Points |
Focus |
| Frontmatter Validity |
25 |
YAML syntax, required fields, correct field names |
| Glob Pattern Quality |
20 |
Syntax, specificity, quoting |
| Content Quality |
25 |
Actionable, concise, non-conflicting |
| Organization |
15 |
Naming, structure, symlink usage |
| Token Efficiency |
15 |
Rule size, redundancy detection |
| Score |
Level |
| 91-100 |
Excellent - Production-ready |
| 76-90 |
Good - Minor improvements possible |
| 51-75 |
Basic - Needs optimization |
| 26-50 |
Below Standards - Significant issues |
| 0-25 |
Critical - Invalid or broken rules |
Resources
Skill-Specific Modules
- Frontmatter Validation: See
modules/frontmatter-validation.md
- Glob Pattern Analysis: See
modules/glob-pattern-analysis.md
- Content Quality Metrics: See
modules/content-quality-metrics.md
- Organization Patterns: See
modules/organization-patterns.md
Tools
- Rules Validator:
scripts/rules_validator.py
Related Skills
abstract:skills-eval - Skill evaluation frameworkabstract:hooks-eval - Hook evaluation framework
Exit Criteria
1---2name: rules-eval-23description: Evaluate Claude Code rules in .claude/rules/. Use for frontmatter, globs, and quality audits.4---5# Rules Evaluation Framework67## When NOT To Use89- Evaluating skills (use `abstract:skills-eval`)10- Evaluating hooks (use `abstract:hooks-eval`)11- Writing a new rule (use `hookify:writing-rules`)1213## Table of Contents14151. [Overview](#overview)162. [Quick Start](#quick-start)173. [Evaluation Workflow](#evaluation-workflow)184. [Scoring](#scoring)195. [Resources](#resources)2021## Overview2223This skill evaluates Claude Code rules in `.claude/rules/` directories against quality standards. It validates YAML frontmatter, glob pattern syntax, content quality, and directory organization. Rules files support path-scoped conditional loading via `paths` frontmatter and unconditional rules (no `paths` field).2425Key validations: YAML syntax errors, unquoted glob patterns, Cursor-specific fields (`alwaysApply`, `globs`), overly broad patterns, content verbosity, and naming conventions.2627## Quick Start2829```bash30# Evaluate rules in current project31/rules-eval3233# Evaluate specific directory34/rules-eval .claude/rules/3536# Detailed analysis with recommendations37/rules-eval --detailed38```3940## Evaluation Workflow41421. Scan `.claude/rules/` for all `.md` files (including subdirectories)432. Validate YAML frontmatter syntax and fields443. Analyze glob patterns for correctness and specificity454. Assess content quality (actionable, concise, non-conflicting)465. Check organization (naming, structure, symlinks)476. Measure token efficiency and redundancy4849## Scoring5051| Category | Points | Focus |52|----------|--------|-------|53| Frontmatter Validity | 25 | YAML syntax, required fields, correct field names |54| Glob Pattern Quality | 20 | Syntax, specificity, quoting |55| Content Quality | 25 | Actionable, concise, non-conflicting |56| Organization | 15 | Naming, structure, symlink usage |57| Token Efficiency | 15 | Rule size, redundancy detection |5859| Score | Level |60|-------|-------|61| 91-100 | Excellent - Production-ready |62| 76-90 | Good - Minor improvements possible |63| 51-75 | Basic - Needs optimization |64| 26-50 | Below Standards - Significant issues |65| 0-25 | Critical - Invalid or broken rules |6667## Resources6869### Skill-Specific Modules70- **Frontmatter Validation**: See `modules/frontmatter-validation.md`71- **Glob Pattern Analysis**: See `modules/glob-pattern-analysis.md`72- **Content Quality Metrics**: See `modules/content-quality-metrics.md`73- **Organization Patterns**: See `modules/organization-patterns.md`7475### Tools76- **Rules Validator**: `scripts/rules_validator.py`7778### Related Skills79- `abstract:skills-eval` - Skill evaluation framework80- `abstract:hooks-eval` - Hook evaluation framework8182## Exit Criteria8384- [ ] Every `.md` file under `.claude/rules/` (including subdirectories) receives a quality score85 (0-100) with per-category breakdown across the five dimensions.86- [ ] YAML frontmatter syntax errors and unquoted glob patterns are listed as blocking findings87 before any score is reported.88- [ ] Rules using Cursor-specific fields (`alwaysApply`, `globs`) are flagged as89 non-compliant with the expected Claude Code schema.90- [ ] Any rule scoring below 26 (Critical tier) is reported with at least one concrete91 corrective action, not just a score.
athola