skill-to-mcp

🥈Silver

Converts AI skills from Claude Skills format to MCP server resources. Enables biomedical researchers and developers to integrate AI skills into their workflows. Connects to BioContextAI's registry and MCP servers for deployment.

2340Updated 1mo ago

Intermediate30min to implementautomation

Saves ~60 min per use

Quick InstallView Source

git clone https://github.com/biocontext-ai/skill-to-mcp.git

Works with:

Claude

Overview

About This Skill

Skill-to-MCP automatically discovers and exposes Claude Skills as accessible MCP server resources through the Model Context Protocol. It recursively finds SKILL.md files, parses YAML frontmatter metadata, and provides three core tools: get_available_skills to list all skills, get_skill_details to retrieve full skill content, and get_skill_related_file to access supporting files. The server includes built-in security with path validation to prevent directory traversal attacks. Designed for biomedical researchers and developers, it integrates with BioContextAI's registry while remaining domain-agnostic for any skill collection.

How to Use

Install via uvx or pip, configure a skills directory containing SKILL.md files with YAML frontmatter, and specify the directory path using --skills-dir or SKILLS_DIR environment variable. Add the server to your MCP client configuration (Claude Desktop, Cursor, etc.) and access skills through the three provided tools: get_available_skills, get_skill_details, and get_skill_related_file.

Use Cases

Expose a collection of biomedical research skills to Claude or other LLM agents

Create an MCP server from existing Claude Skills documentation

Manage multiple skill collections for different research teams or projects

Enable Claude Desktop or other MCP clients to access structured skill libraries

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/biocontext-ai/skill-to-mcp

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Convert the following AI skill from Claude Skills format to MCP server resources: [SKILL_NAME]. Ensure compatibility with BioContextAI's registry and MCP servers. Include all necessary metadata and dependencies. [SKILL_DESCRIPTION]

Example Output

# MCP Server Resource Conversion

## Skill Name: Drug Interaction Analyzer

## Description
Converts and analyzes potential drug interactions based on patient data and medical history.

## Metadata
- **Category**: Biomedical Analysis
- **Input Type**: Text (Patient Data, Medical History)
- **Output Type**: Text (Drug Interaction Analysis)
- **Dependencies**: BioContextAI Registry, MCP Servers

## Conversion Details
- **Input Format**: JSON
- **Output Format**: JSON
- **API Endpoint**: `/api/v1/drug_interaction_analyzer`

## Example Input
```json
{
  "patient_data": {
    "age": 65,
    "gender": "male",
    "medical_history": ["hypertension", "diabetes"],
    "current_medications": ["lisinopril", "metformin"]
  }
}
```

## Example Output
```json
{
  "interactions": [
    {
      "drug1": "lisinopril",
      "drug2": "metformin",
      "severity": "low",
      "description": "No significant interactions expected."
    },
    {
      "drug1": "lisinopril",
      "drug2": "aspirin",
      "severity": "moderate",
      "description": "Monitor for signs of hyperkalemia."
    }
  ]
}
```