skill_to_mcp

skill_to_mcp#

Submodules#

Attributes#

mcp

__version__

Functions#

run_app([transport, port, hostname, environment, ...])

Run the MCP server "skill-to-mcp".

Package Contents#

skill_to_mcp.run_app(transport: str = 'stdio', port: int = 8000, hostname: str = '0.0.0.0', environment: str = 'development', version: bool = False, skills_dir: str | None = None)#

Run the MCP server “skill-to-mcp”.

Convert AI Skills (following Claude Skills format) into MCP server resources. This server exposes skills as MCP tools, allowing LLM applications to discover and access specialized knowledge and workflows stored in SKILL.md files.

Configuration options: - Skills directory: Set via “-s/–skills-dir” or SKILLS_DIR environment variable

If not provided, uses default help skill with usage instructions

  • Transport: Set via “-t/–transport” or MCP_TRANSPORT environment variable (default: “stdio”)

  • Port: Set via “-p/–port” or MCP_PORT environment variable (default: 8000)

  • Hostname: Set via “-h/–host” or MCP_HOSTNAME environment variable (default: “0.0.0.0”)

  • Environment: Set via “-e/–env” or MCP_ENVIRONMENT environment variable (default: “development”)

skill_to_mcp.mcp: fastmcp.FastMCP | None = None#
skill_to_mcp.__version__#
Contents