faf_bench

MCP tool from Claude Faf MCP by Wolfe-Jam

Benchmark AI grounding — cold vs .faf, with a ✪ receipt

How to use it

faf_bench is exposed by the Claude Faf MCP MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the faf_bench tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.

Install Claude Faf MCP

$npx faf-cli auto

FULL CLAUDE FAF MCP LISTING

Other tools in Claude Faf MCP (22)

What is .faf?

Import/export AGENTS.md (OpenAI Codex)

Auto-detect stack and populate context

Validate .faf structure

Import/export Conductor directory

Import/export .cursorrules (Cursor IDE)

Environment inspection

Diagnose and fix common issues

Discover all formats in your project

Import/export GEMINI.md (Google Gemini)

Extract context from any GitHub repo URL

Guided interview to Gold Code

Add human context (the 6Ws)

Initialize project DNA

Lightning-fast creation (3ms)

Read any file

Extract context from README (+25-35% boost)

AI-readiness score (0-100%) with breakdown

Project status overview

Sync .faf → CLAUDE.md — agents/cursor/gemini/copilot/all also emit AGENTS.md / .cursorrules / GEMINI.md / copilot-instructions.md

Tri-sync .faf ↔ CLAUDE.md ↔ MEMORY.md — Pro feature, free for developers 🐘

Write any file