FAF

Provides a universal MCP server for real-time AI context synchronization across codebases and MCP-compatible environments.

javascript

2

GitHub Stars

javascript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "wolfe-jam-faf-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "faf-mcp"
      ]
    }
  }
}

FAF-MCP provides a universal MCP server that keeps your AI context aligned across your codebase and development tools. It enables consistent, low-drift integration with Cursor, Windsurf, Cline, VS Code, Claude Desktop, and other MCP-compatible environments, so your AI assistants stay informed and productive across sessions and platforms.

How to use

You will run FAF-MCP as a local or remote MCP server and connect your MCP-enabled client to it. Use the unified .faf workflow to keep context synchronized across platforms, so your AI partner remains aware of your project’s history, architecture, and team context. Start prompts with the tool-enabled pattern to initialize, score readiness, or sync context across environments.

How to install

Prerequisites: you need Node.js and npm installed on your system.

npm install -g faf-mcp

Next, add FAF-MCP to your MCP configuration using the standard stdio approach.

{"mcpServers": {"faf": {"command": "npx", "args": ["-y", "faf-mcp"]}}}

Platform-specific setup often places the MCP config in a platform-defined location. For example, you may find the following paths in your environment after installation.

Additional content

Eternal bi-sync keeps FAF context files in perfect alignment: any update to project.faf or CLAUDE.md synchronizes in milliseconds, eliminating drift and manual maintenance.

The system uses a tiered readiness model to indicate how optimized your AI is for the project. At 100%, AI is fully optimized and strongly aligned with project DNA, architecture, and team context.

A set of native MCP tools is available to manage initialization, scoring, syncing, enhancements, and file parsing. You can invoke them directly from the CLI or through the configured MCP client.

CLI prompts and usage patterns

Start your prompts with the trigger phrase to invoke FAF MCP tools. Examples include scoring AI readiness, syncing context, and enhancing the project in place.

Ecosystem and license

The FAF-MCP ecosystem includes multiple components such as claude-faf-mcp, faf-cli, faf-wasm, and the official faf.one site. It is released under the MIT License and designed for broad compatibility across MCP-enabled platforms.

Available tools

faf_init

Initialize a new project FAF context file and prepare it for synchronization across platforms.

faf_score

Assess AI-readiness from 0 to 100%, guiding optimization efforts and indicating drift risk.

faf_sync

Synchronize context across all connected MCP platforms to keep information aligned.

faf_bi_sync

Enable bi-directional synchronization between FAF context files and CLAUDE.md.

faf_enhance

Apply intelligent enhancements to FAF context for improved AI assistance.

faf_read

Parse and validate FAF context files to ensure correct formatting and consistency.

faf_write

Create or update FAF context with validation to maintain integrity across sessions.