mcp-agentskills 0.1.0

An MCP server for bringing Anthropic's Agent Skills to any MCP-compatible agent.

AgentSkills MCP: Bringing Anthropic's Agent Skills to Any MCP-compatible Agent

简体中文 | English

📖 Project Overview

Agent Skills is a new function recently introduced by Anthropic. By packaging specialized skills into modular resources, it allows Claude to transform on demand into a “tailored expert” suited to any scenario. AgentSkills MCP, built on the FlowLLM framework, unlocks Claude’s proprietary Agent Skills for any MCP-compatible agent. It implements the Progressive Disclosure architecture proposed in Anthropic’s official Agent Skills engineering blog, enabling agents to load necessary skills as needed, thereby efficiently utilizing limited context windows.

💡 Why Choose AgentSkills MCP?

• ✅ Zero-Code Configuration: one-command install (pip install agentskills-mcp)
• ✅ Out-of-the-Box: uses official Skill format and fully compatible with Anthropic’s Agent Skills
• ✅ MCP Support: multiple transports (stdio/SSE/HTTP), works with any MCP-compatible agent
• ✅ Flexible Skill Path: custom skill directories with automatic detection, parsing, and loading

🔥 Latest Updates

• [2025-12] 🎉 Released agentskills-mcp v0.1.0

🚀 Quick Start

Installation

Install AgentSkills MCP with pip:

pip install agentskills-mcp

Or with uv:

uv pip install agentskills-mcp

For Development (if you want to modify the code):

git clone https://github.com/zouyingcao/agentskills-mcp.git
cd agentskills-mcp

conda create -n agentskills-mcp python==3.10
conda activate agentskills-mcp
pip install -e .

---

Load Skills

1. Create a directory to store Skills, like:

mkdir skills

2. Clone from open-source GitHub repositories, e.g.,

https://github.com/anthropics/skills
https://github.com/ComposioHQ/awesome-claude-skills

3. Add the collected Skills into the directory created in step 1. Each Skill is a folder containing a SKILL.md file.

---

Run

Local process communication (stdio)

_{This mode runs AgentSkills MCP via uvx and communicates through stdin/stdout, suitable for local MCP clients.}

{
  "mcpServers": {
    "agentskills-mcp": {
      "command": "uvx",
      "args": [
        "agentskills-mcp",
        "config=default",
        "mcp.transport=stdio",
        "metadata.skill_dir=\"./skills\""
      ],
      "env": {
        "FLOW_LLM_API_KEY": "xxx",
        "FLOW_LLM_BASE_URL": "https://dashscope.aliyuncs.com/compatible-mode/v1"
      }
    }
  }
}

Remote communication (SSE/HTTP Server)

_{This mode runs AgentSkills MCP as a standalone SSE/HTTP server that can be accessed remotely.}

- Step 1: Configure Environment Variables

Copy example.env to .env and fill in your API key:

cp example.env .env
# Edit the .env file and fill in your API key

- Step 2: Start the Server

Start the AgentSkills MCP server with SSE transport:

agentskills-mcp \
  config=default \
  mcp.transport=sse \
  mcp.host=0.0.0.0 \
  mcp.port=8001 \
  metadata.skill_dir="./skills"

The service will be available at: http://0.0.0.0:8001/sse

- Step 3: Connect from MCP Client

• Add this configuration to your MCP client (Cursor, Gemini Code, Cline, etc.) to connect to the remote SSE server:

{
  "mcpServers": {
    "agentskills-mcp": {
      "type": "sse",
      "url": "http://0.0.0.0:8001/sse"
    }
  }
}

• You can also use the FastMCP Python client to directly access the server:

import asyncio
from fastmcp import Client


async def main():
    async with Client("http://0.0.0.0:8001/sse") as client:
        tools = await client.list_tools()
        for tool in tools:
            print(tool)

        result = await client.call_tool(
            name="load_skill",
            arguments={
              "skill_name"="pdf"
            }
        )
        print(result)


asyncio.run(main())

One-Command Test

_{This command will start the server, connect via FastMCP client, and test all available tools automatically.}

python tests/run_project_sse.py <path/to/skills>
or
python tests/run_project_http.py <path/to/skills>

Demo

After starting the AgentSkills MCP server with the SSE transport, you can run the demo:

# Enable Agent Skills for the Qwen model.
# Since Qwen supports function calling, you can implement Agent Skills by passing the MCP tools registered by the AgentSkills MCP service to the tools parameter.
cd tests
python run_skill_agent.py

---

🔧 MCP Tools

This service provides four tools to support Agent Skills:

• load_skill_metadata_op — Loads the names and descriptions of all Skills into the agent context at startup (always called)
• load_skill_op — When a specific skill is needed, loads the SKILL.md content by skill name (invoked when triggering the Skill)
• read_reference_file_op — Reads specific files from a skill, such as scripts or reference documents (on demand)
• run_shell_command_op — Executes shell commands to run executable scripts included in the skill (on demand)

For detailed parameters and usage examples, see the documentation.

⚙️ Server Configuration Parameters

Parameter	Description	Example
config	Configuration files to load (comma-separated). Default: default (core workflow)	config=default
mcp.transport	Transport mode: stdio (stdin/stdout, good for local), sse (Server-Sent Events, good for online apps), http (RESTful, good for lightweight remote calls)	mcp.transport=stdio
mcp.host	Host address (for sse/http transport only)	mcp.host=0.0.0.0
mcp.port	Port number (for sse/http transport only)	mcp.port=8001
metadata.skill_dir	Skills Directory (required)	metadata.dir=./skills

For the full set of available options and defaults, refer to default.yaml.

Environment Variables

Variable Name	Required	Description
FLOW_LLM_API_KEY	✅ Yes	API key for OpenAI-compatible LLM Service
FLOW_LLM_BASE_URL	✅ Yes	Base URL for OpenAI-compatible LLM Service

---

🤝 Contributing

We welcome community contributions! To get started:

1. Install the package in development mode:

pip install -e .

2. Install pre-commit hooks:

pip install pre-commit
pre-commit run --all-files

3. Submit a pull request with your changes.

---

📚 Learn More

• Anthropic Agent Skills Documentation
• Anthropic Engineering Blog
• Claude Agent Skills: A First Principles Deep Dive
• FlowLLM Documentation
• MCP Documentation

⚖️ License

This project is licensed under the Apache License 2.0 — see LICENSE for details.

---

📈 Star History