# .FAF Context

> .FAF Context is an open-source MCP server for developer tools. Writes AGENTS.md, CLAUDE.md and .cursorrules from one project.faf file, locally. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, vscode, self-hosted
- **Website:** https://faf.one
- **Repository:** https://github.com/wolfe-jam/faf-mcp
- **MCP Registry name:** one.faf/faf-mcp
- **Categories:** developer tools, knowledge and memory
- **Listing updated:** 2026-11-14

## About

FAF keeps a project's context in a single YAML file, project.faf, and generates each tool's own file from it: AGENTS.md, .cursorrules, GEMINI.md and CLAUDE.md. Each is scored so you can see where the context is thin. This server runs locally over stdio on the same faf-cli the terminal uses.

Version 4.2.0 changed who fills in the human slots: faf_go shows each suggested goal or answer with its source, such as the README or package.json, and only what you confirm is written. faf_init and faf_auto leave them empty and report how many suggestions wait, so a new file scores honestly low. faf_write refuses to overwrite existing context files without overwrite: true and never touches the .faf-dna birth certificate.

The project is MIT-licensed and the media type is registered with IANA.

## Features

- **One source file:** project.faf drives every tool's context file.
- **Confirm before writing:** Suggested goals are shown with sources; only confirmed ones are saved.
- **Context scoring:** Scores show what is missing.
- **Overwrite guards:** Existing files need explicit overwrite.
- **Local stdio:** Runs on your machine via npm.

## Good for

- Keep CLAUDE.md and .cursorrules in sync
- Bootstrap context for a new repo
- Find gaps in project context
- Give an editor's AI a project summary

## Install

### Claude Code (Terminal)

```sh
claude mcp add faf-mcp -- npx -y faf-mcp
```

### Claude Desktop (claude_desktop_config.json)

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

### Cursor (~/.cursor/mcp.json)

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

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "faf-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "faf-mcp"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

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

### Other clients (your client's MCP settings (most use this shape))

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

## Tools

- `faf_go`: Offers sourced suggestions for goal and answers; writes only confirmed ones.
- `faf_init`: Creates a new project.faf.
- `faf_auto`: Automatic setup leaving human slots empty.
- `faf_readme`: Suggests README content without writing.
- `faf_sync`: Updates name and stack from package.json.
- `faf_claude`: Writes CLAUDE.md and chained files.
- `faf_write`: Writes files with overwrite guards.
- `faf_dna`: Handles the birth certificate.
- `faf_git`: Writes the goal from the GitHub description.
- `faf_doctor`: Explains why context is incomplete.
- `faf_score`: Shows what is missing.

## FAQ

### Which files does it generate?

AGENTS.md, .cursorrules, GEMINI.md and CLAUDE.md.

### Does the AI write my project goal?

Not since 4.2.0; suggestions are shown and only your confirmed answers are written.

### How do I install it?

npm package faf-mcp, stdio.

### Licence?

MIT.

### Is .FAF Context free?

Yes. .FAF Context is open source, so you can use and self-host it for free.

### How do I add .FAF Context to Claude Code?

Run this command in your terminal: claude mcp add faf-mcp -- npx -y faf-mcp

### What are the best alternatives to .FAF Context?

The closest alternatives to .FAF Context are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Nautilus Compass Memory](https://ai-powered.com/mcp/nautilus-compass-memory): Nautilus Compass Memory is an open-source MCP server for knowledge and memory.

---
Source: https://ai-powered.com/mcp/faf-context
