skill-refiner

What This Skill Does

The skill-refiner is an essential maintenance tool for OpenClaw users, designed to automate the audit and remediation of AI agent skills throughout your workspace. It performs a deep-scan of your local directories to identify every SKILL.md file, ensuring they strictly adhere to the unified compliance standards defined by the OpenClaw skill-creator framework. By programmatically detecting issues like malformed YAML frontmatter, incorrect directory naming conventions, and the presence of extraneous documentation files, this skill maintains the integrity and performance of your agent library. It doesn't just identify problems; it systematically fixes them by enforcing strict naming, relocating orphaned skills to the designated workspace directory, and sanitizing repository structures.

Installation

You can install this skill directly via the command line within your OpenClaw environment. Execute the following command: clawhub install openclaw/skills/skills/1va7/skill-refiner Once installed, the refiner will be available to monitor your workspace and can be triggered whenever your skill library starts to drift from established standards.

Use Cases

• Workspace Cleanup: Use this after importing several new community skills to ensure they follow your local standards.
• Migration Assistance: If you are reorganizing your workspace structure, the refiner can identify and relocate skills that have been displaced from the standard directory.
• Quality Assurance: Automated verification that all skills possess valid frontmatter and accurate trigger descriptions, ensuring the OpenClaw agent can correctly interpret and execute them.

Example Prompts

• "Refine all skills in the current workspace and fix any non-compliant configurations found."
• "Run a full quality audit on my skill library and generate a report of any issues."
• "Audit my skills and fix any missing or malformed frontmatter automatically."

Tips & Limitations

• Permissioning: The skill requires file-write access as it actively modifies and deletes files during the repair process. Always back up your custom skills before running a bulk audit.
• Scope: While it searches the entire workspace, it assumes a standard file structure for scripts, assets, and references. Custom, non-standard project structures may require manual review.
• Performance: For massive workspaces containing thousands of files, the initial scan may take a moment. Use this periodically rather than after every minor file change.