Back to Skills

rule-making-skill

Analyze a specific directory (e.g. frontend/, backend/, e2e/) and generate .claude/rules/ markdown files for it. Use when asked to create rules, analyze a folder for Claude Code, or set up domain memory for a specific part of the codebase.

5stars1forksUpdated 2/21/2026

Security Assessment

Safe(100/100)
Security Score100/100

About rule-making-skill

The 'rule-making-skill' is designed to automate the process of analyzing specific directories within a codebase, such as 'frontend/', 'backend/', or 'e2e/', and generating relevant markdown files for Claude Code domain memory. This skill is aimed at solving the problem of maintaining structured, persistent knowledge about the codebase, enabling more efficient development and decision-making. By transforming information into rule files, it ensures that agents can work with a grounded sense of context during their operations, avoiding the limitations of generalized agents that start each session without prior context.

The skill works by scanning the target directories, detecting key concerns based on a predefined matrix of patterns, and then creating markdown files (.claude/rules/) that serve as quick-reference documents. These files include essential domain knowledge such as goals, current state, and system scaffolding, which are essential for understanding and managing the codebase. It integrates with existing task boards and directories to provide up-to-date, actionable insights, which are crucial for tasks like setting up domain memory or creating rules based on code analysis.

Target users include developers, DevOps teams, and anyone managing complex codebases that require clear and structured documentation for Claude agents. This tool is particularly useful for teams looking to ensure that their domain memory is properly initialized and aligned with their codebase structure. It's also valuable for automating the creation of rules that would otherwise require extensive manual effort.

FAQ

How do I use the 'rule-making-skill'?

Simply specify the target directory (e.g., 'frontend/', 'backend/', or 'e2e/') and request the generation of rules or domain memory for that part of the codebase. The skill will analyze the directory and generate relevant markdown rule files.

Does this skill work with any directory?

Yes, the skill can be used with any directory within the codebase, provided it contains concerns that match the predefined patterns (e.g., 'auth', 'data', 'api'). The skill scans the directory and generates rule files based on the detected concerns.

What kind of concerns does the skill detect?

The skill detects a variety of concerns such as authentication, data handling, API routes, validation, state management, components, and many more. These concerns are identified using specific patterns, such as file names, function names, or imported modules.

Are there any prerequisites for using this skill?

The skill requires access to the codebase's CLAUDE.md files and task boards. It works best when task-related documentation and directory structures are already in place.

Can the skill generate rules for custom concerns?

The skill is currently designed to detect and generate rules for a set of predefined concerns. Custom concerns would require modifying the detection matrix or extending the skill.

Install rule-making-skill

Download and extract the skill files to your .claude/skills/ directory.

Quick Setup:

  1. Copy the skill folder to .claude/skills/
  2. Claude will automatically detect and use the skill