skill-refiner for Openclaw

An automated auditing and repair tool designed to ensure every skill in your workspace complies with strict quality and structural requirements.

1va7
v1.0.0
Feb 20, 2026
0
1.8k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install skill-refiner

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install skill-refiner using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is skill-refiner?

The skill-refiner is a specialized maintenance tool built to scan your entire development environment for Openclaw Skills. It identifies SKILL.md files that deviate from established standards, checking for proper YAML frontmatter, naming conventions, and directory organization.

This tool is essential for maintaining a high-quality agent library, as it provides an automated pathway to repair non-compliant skills. It handles everything from correcting malformed metadata to relocating skills into the standard directory structure, ensuring your Openclaw Skills remain clean, searchable, and fully functional.

skill-refiner Use Cases

  • Auditing a large workspace to find and fix non-compliant Openclaw Skills.
  • Automatically correcting malformed YAML frontmatter in SKILL.md files.
  • Relocating misplaced skills from temporary folders to the official skills directory.
  • Standardizing skill naming conventions to lowercase-hyphen-case.
  • Cleaning up workspace clutter by removing redundant files like README.md or CHANGELOG.md.

How skill-refiner Works

  1. Search the entire workspace recursively to discover every instance of a SKILL.md file.
  2. Run a Python-based audit on each discovered skill to detect blocking issues or advisory warnings.
  3. Generate a summarized report in a table format to visualize the compliance status of all skills.
  4. Apply automated remediation steps to fix frontmatter, descriptions, and file structures.
  5. Re-run the audit to validate that all Openclaw Skills now meet the compliance checklist requirements.

skill-refiner Setup

To begin auditing your Openclaw Skills, ensure you have the necessary audit scripts in your workspace and execute them via the CLI.

# Discover all skills in the current workspace
bash scripts/find_skills.sh .

# Audit a specific skill for compliance issues
python3 scripts/audit_skill.py <skill-dir>

skill-refiner Data Schema & Taxonomy

The skill-refiner enforces a strict organization for Openclaw Skills to ensure compatibility and performance.

Component Requirement
YAML Frontmatter Must contain exactly name and description (no extra fields)
Directory Naming lowercase-hyphen-case only, maximum 64 characters
File Structure SKILL.md must be at the root; no README.md or CHANGELOG.md allowed
Resource Folders All assets must reside in scripts/, references/, or assets/ folders
Body Length SKILL.md content should be under 500 lines to maintain focus

skill-refiner Advanced Features

  • Exhaustive workspace scanning that finds Openclaw Skills even when they are not in the default directory.
  • Automated remediation for weak descriptions, adding necessary trigger language automatically.
  • Intelligent file extraction that moves long documentation sections into the references folder while maintaining links.
  • Batch processing capabilities to audit and fix hundreds of Openclaw Skills in a single workflow.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*