MCP Catalogs
Home

fabric-mcp

by ksylvan·86·Score 47

Fabric MCP Server integrates Daniel Miessler's Fabric AI framework into MCP-enabled tools through a standalone server.

ai-llmdeveloper-toolsproductivity
13
Forks
2
Open issues
this month
Last commit
2d ago
Indexed

Overview

This is a full-featured MCP server that bridges Fabric's AI capabilities with the Model Context Protocol. It allows users to access Fabric's specialized prompts and patterns directly within MCP-enabled environments like IDEs and chat interfaces. The server translates MCP requests into REST API calls to a running Fabric instance, enabling seamless workflow integration without context switching.

Try asking AI

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

you:Using Fabric's specialized prompts for code explanation and refactoring directly within IDEs
you:Accessing Fabric's pattern library through MCP-enabled chat interfaces
you:Leveraging Fabric's AI capabilities in custom MCP-based applications
you:What is the relationship between Fabric MCP Server and the Fabric framework?
you:Does this server require running a separate Fabric instance?

When to choose this

Choose this MCP server when you're already using Fabric's prompt patterns and want to integrate them directly into MCP-compatible tools like IDEs and chat interfaces.

When NOT to choose this

Don't choose this if you're not already using Fabric, as it requires a separate Fabric instance running with the --serve flag.

Tools this server exposes

6 tools extracted from the README
  • fabric_run_pattern

    Execute a Fabric pattern with the provided input.

  • list_patterns

    Retrieve a list of available Fabric patterns.

  • get_pattern_details

    Get detailed information about a specific Fabric pattern.

  • list_models

    List available AI models in Fabric.

  • list_strategies

    Retrieve available strategies in Fabric.

  • get_configuration

    Retrieve the current Fabric configuration.

Comparable tools

fabric-apiprompt-engineering-mcpllm-tool-integration

Installation

Installation

From PyPI (for users)

pip install fabric-mcp

From Source (for developers)

git clone https://github.com/ksylvan/fabric-mcp.git
cd fabric-mcp
uv sync --dev
source .venv/bin/activate

Configuration

Set environment variables:

  • FABRIC_BASE_URL: Base URL of Fabric REST API (default: http://127.0.0.1:8080)
  • FABRIC_API_KEY: API key for authentication (if required)

Running the Server

# Standard I/O transport (default)
fabric-mcp --stdio

# HTTP transport
fabric-mcp --http-streamable --host 0.0.0.0 --port 3000

Claude Desktop Configuration

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "fabric": {
      "command": "fabric-mcp",
      "args": ["--stdio"]
    }
  }
}

FAQ

What is the relationship between Fabric MCP Server and the Fabric framework?
The Fabric MCP Server is a standalone implementation that acts as a bridge between MCP clients and the Fabric framework. It translates MCP requests into REST API calls to a running Fabric instance.
Does this server require running a separate Fabric instance?
Yes, the Fabric MCP Server requires a running Fabric instance with the `fabric --serve` command exposing its REST API. The server connects to this instance to process requests.

Compare fabric-mcp with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.