augmentedivan/fabric-mcp-server

Table of Contents
1. Introduction
2. What is Model Context Protocol (MCP)?
3. Features
4. Tools
5. Installation
6. Usage
7. Configuration for Claude Desktop
8. Configuration for VS Code with Cline
9. Tips for Using with Different AI Agents
10. Troubleshooting
11. Contributing
12. License
Introduction
The fabric-mcp-server is a Model Context Protocol (MCP) server designed to expose Daniel Miessler's Fabric patterns as tools for integration with AI coding agents and assistants. This integration enhances AI capabilities by leveraging AI-driven pattern execution from the Fabric repository. The server works with various AI platforms including Claude Desktop, Cline, and other MCP-compatible AI agents.
What is Model Context Protocol (MCP)?
The Model Context Protocol (MCP) is a specification that facilitates communication between AI systems and external tools or resources. It standardizes the way AI models interact with various capabilities such as databases, APIs, and file systems. MCP servers, like fabric-mcp-server, implement this protocol to make tools and resources accessible to AI models, thereby expanding their functional scope.
Features
- **Exposes Fabric Patterns as Tools**: The server makes all Fabric patterns available as individual tools within MCP-compatible AI agents.
- **Pattern Execution**: Users can select and execute Fabric patterns directly within AI assistant tasks.
- **Enhanced Capabilities**: Integrates AI-driven pattern execution to augment AI assistant functionality.
- **Cross-Platform Compatibility**: Works with Claude Desktop, Cline, and other MCP-compatible AI agents.
Tools
The fabric-mcp-server exposes a wide range of Fabric patterns as tools. Some examples include:
- analyze_claims
- summarize
- extract_wisdom
- create_mermaid_visualization
- And many more...
To see the full list of available patterns, you can list the directories in the fabric/patterns directory.
Installation
1. **Clone the Repository**: Clone the fabric-mcp-server repository to your local system.
2. **Install Dependencies**: Navigate into the fabric-mcp-server directory and run npm install.
3. **Build the Project**: Run npm run build to compile the TypeScript code.
Usage
To use the fabric-mcp-server with AI agents:
1. Ensure the server is installed and running.
2. Configure the MCP server in your AI agent's settings file.
3. Create a new task or conversation and select a Fabric pattern to use.
The specific configuration steps vary depending on which AI agent you're using. See the sections below for detailed instructions.
Configuration for Claude Desktop
To use fabric-mcp-server with Claude Desktop:
1. **Complete Installation**: Follow the installation steps above to build the project.
2. **Configure Claude Desktop**: Add the MCP server configuration to your Claude Desktop settings. The configuration file is typically located at:
- **Windows**: %APPDATA%\Claude\claude_desktop_config.json
- **macOS**: ~/Library/Application Support/Claude/claude_desktop_config.json
- **Linux**: ~/.config/Claude/claude_desktop_config.json
3. **Add Server Configuration**: Add the following to your claude_desktop_config.json: