rulecheck

Scan codebases for CLAUDE.md violations, apply fixes, and create pull requests.

Updated May 21, 2026
One-click install
npx skills add https://github.com/sunzcdev/archon-dev-team --skill rulecheck-sunzcdev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rulecheck
Source: https://github.com/sunzcdev/archon-dev-team/tree/main/.claude/skills/rulecheck
Command: npx skills add https://github.com/sunzcdev/archon-dev-team --skill rulecheck-sunzcdev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

CLAUDE.md-based engineering rules are often inconsistently followed, so AI-driven code changes can violate architectural, safety, logging, and error-handling conventions without anyone catching them.

Core Features & Use Cases

  • Autonomous rule scanning and fixing: Delegates to a dedicated rulecheck agent that finds CLAUDE.md violations, applies fixes, and validates them in an isolated worktree.
  • Validation and PR creation workflow: Runs full validation, commits changes, and creates a pull request with the outcomes.
  • Progress memory across runs: Uses persistent project memory so future runs prioritize what was previously fixed and what remains.

Quick Start

Ask an AI to run the rulecheck skill with an optional focus area, for example: Invoke /rulecheck error handling to scan and fix the highest-impact CLAUDE.md error-handling violations.

Frequently Asked Questions about rulecheck

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

FAQPage Schema
How do I autonomously fix CLAUDE.md rule violations in my codebase?

To autonomously fix CLAUDE.md rule violations, use a dedicated rulecheck agent that scans the codebase, applies fixes for high-impact violations, and validates them in an isolated worktree before creating a pull request.

How does persistent memory track codebase scanning progress across multiple runs?

Persistent memory tracks codebase scanning progress by storing previous fix outcomes in project memory, allowing future autonomous agent runs to prioritize remaining violations and avoid re-processing resolved issues.

Can I run autonomous fixes for lint validation and rule compliance without affecting my main branch?

Yes, you can run autonomous fixes for lint validation and rule compliance safely by using worktree isolation, which executes code changes and validation in a dedicated environment separate from your main branch.

What is the best way to automate pull request creation after fixing architectural and error-handling conventions?

The best way to automate pull request creation after fixing conventions is to use a safety-gated execution environment that runs full lint validation, commits the changes, and creates a PR with the outcomes automatically.

Does rule check work with specific focus areas like error handling during codebase scanning?

Yes, rule check works with specific focus areas like error handling; you can invoke the skill with an optional focus parameter to scan and fix the highest-impact CLAUDE.md violations within that targeted domain.

Why does my AI-driven code change violate architectural conventions despite having defined rules?

AI-driven code changes violate architectural conventions because rules are often inconsistently followed, requiring a repeatable governance workflow with safety-gated execution and autonomous scanning to enforce compliance.