skill-to-mcp

Overview

Skill-to-MCP is a Python package that transforms Claude Skills format into MCP server resources, enabling LLM applications to access skill collections through the Model Context Protocol. The server automatically discovers skills in specified directories, parses YAML frontmatter from SKILL.md files, and provides three core tools: get_available_skills for listing all skills, get_skill_details for retrieving content and file listings, and get_skill_related_file for accessing specific files with security protections against directory traversal.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• get_available_skills

  Lists all available skills with metadata

  Try:Show me all available skillsTry:List skills in the repository
• get_skill_details

  Returns SKILL.md content and file listing for a specific skill

  Try:Get details for the single-cell-rna-qc skillTry:Show me the content of the analysis skill
• get_skill_related_file

  Reads a specific file within a skill directory

  Try:Read the qc_analysis.py script from the single-cell-rna-qc skillTry:Show me the guidelines.md file in the references folder

Comparable tools

Installation

pip install --user skill_to_mcp

SKILLS_DIR=/path/to/skills uvx skill_to_mcp

{
  "mcpServers": {
    "skill-to-mcp": {
      "command": "uvx",
      "args": ["skill_to_mcp", "--skills-dir", "/path/to/your/skills"],
      "env": {
        "UV_PYTHON": "3.12"
      }
    }
  }
}

FAQ

What is the format of a skill?

A skill must be in its own subdirectory containing a SKILL.md file with YAML frontmatter specifying name and description, followed by the skill content.

Can I run multiple skill collections simultaneously?

Yes, you can configure multiple instances of the MCP server with different skill directories in your MCP client configuration.

Compare skill-to-mcp with