convez.ai
MCP ServerProdutividadeIntermediário
Wolfe-Jam

Gemini FAF MCP

Wolfe-Jam/gemini-faf-mcp

Contexto persistente de projeto para Gemini — defina uma vez, sincronize sempre

Ver no GitHub
BomPluginScore · de 0 a 100

Servidor MCP que permite ao Google Gemini ler e manter contexto persistente de projetos via arquivos .faf estruturados. Elimina a necessidade de re-explicar stack, convenções e objetivos a cada nova sessão, sincronizando o contexto do projeto automaticamente.

Instalação

$npx convez-ai Wolfe-Jam/gemini-faf-mcp

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
2
Forks
0
Acompanhando
0
Saúde
71%
Último commit
há 2 dias
Versão
v3.3.3
Issues abertas
0
Contribuidores
3

Casos de uso

  • Manter contexto consistente entre múltiplas sessões Gemini
  • Documentar convenções, stack e arquitetura do projeto em formato estruturado
  • Gerar código que respeita padrões e estilos específicos do projeto

Veredito

Solução bem construída que resolve um problema real de sessões stateless com IA. O formato FAF é registrado na IANA e a integração com Cloud Run oferece flexibilidade, mas requer configuração inicial e assume uso regular do Gemini CLI.

README

A tradução para o português deste README ainda está na fila.

gemini-faf-mcp — The Chameleon Edition

Persistent Project Context for Google Gemini. Define once. Sync everywhere.

FAF defines. MD instructs. AI codes.

Stop re-explaining your project to every new Gemini session. Every Gemini conversation starts cold — you re-state your stack, your goals, your conventions every single time. .faf is one structured file that captures all of it. This package is the MCP server that lets Gemini read it.

PyPI FAF Trophy 100% Tests IANA: vnd.faf+yaml IANA: vnd.fafm+yaml DOI: Context paper DOI: Memory paper

Before and after

Without FAF                           With FAF (.faf at 85%+ Bronze)
─────────────────────────             ─────────────────────────
You: "I'm using FastAPI with...       You: "Add a /users/me endpoint"
      PostgreSQL, pytest, and..."     Gemini: [generates correct code,
Gemini: "Got it. What's the              uses your auth pattern,
        codebase like?"                  matches your test style]
You: "It's a REST API for..."
[5 minutes of re-explaining]
Gemini: [now ready to help]

.faf is read once at session start. Every tool call lands on a Gemini that already knows your project.

What's New in v2.4.3

faf_agents and faf_gemini now enhance your files. The structured .faf block is prefixed to the top of AGENTS.md / GEMINI.md for instant AI context, and your Markdown stays in the instruction lane. Re-runs update the block in place — everything you've written below is preserved.

v2.4.2 — The Confinement Edition confined every caller path argument (security: reads restricted to .faf/.fafm, writes to the project root — closes an arbitrary local-file read + faf_init write). v2.4.0 — The Chameleon Edition auto-selects its transport: stdio locally, Streamable HTTP on Cloud Run. Same binary, 12 tools, zero config.


One-Minute Setup

1. Install

pip install gemini-faf-mcp

2. Add to Gemini CLI

gemini extensions install https://github.com/Wolfe-Jam/gemini-faf-mcp

3. Generate your project context

In your Gemini CLI:

> /faf:setup

You should see: Created project.faf — Score: 85% (BRONZE). From this point, every Gemini session in this project reads it automatically.

Tip: A score of 85% (BRONZE) is the minimum where Gemini stops guessing. Run /faf:score to see what's missing and how to push to 100% (TROPHY).


The "One-File" Advantage

A .faf file is structured YAML that captures your project DNA. Every AI agent reads it once and knows exactly what you're building.

# project.faf — your project, machine-readable
faf_version: '2.5.0'
project:
  name: my-api
  goal: REST API for user management
  main_language: Python
stack:
  backend: FastAPI
  database: PostgreSQL
  testing: pytest
human_context:
  who: Backend developers
  what: User CRUD with auth
  why: Replace legacy PHP service

Result: Gemini reads this once and knows your project. No 20-minute onboarding. No wrong assumptions. Every session starts aligned.

FAF defines. MD instructs. AI codes.

What about my GEMINI.md?

You don't replace it. .faf generates it. Run faf_gemini and you get a fresh GEMINI.md with the structured project data baked in as YAML frontmatter — the same GEMINI.md Gemini CLI already reads, but generated from a single source of truth instead of hand-maintained.

> /faf:export
# Generates GEMINI.md from project.faf

.faf is the source. GEMINI.md is one of its outputs. Same logic for AGENTS.md (OpenAI Codex), .cursorrules, CLAUDE.md, and others — write once, render everywhere.


Auto-Detect Your Stack

faf_auto scans your project's manifest files and generates a .faf with accurate slot values. No manual entry needed.

> Auto-detect my project stack
{
  "detected": {
    "main_language": "Python",
    "package_manager": "pip",
    "build_tool": "setuptools",
    "framework": "FastMCP",
    "api_type": "MCP",
    "database": "BigQuery"
  },
  "score": 100,
  "tier": "TROPHY"
}

What it scans:

FileDetects
pyproject.tomlPython + build system + frameworks (FastAPI, Django, Flask, FastMCP) + databases
package.jsonJavaScript/TypeScript + frameworks (React, Vue, Next.js, Express)
Cargo.tomlRust + cargo + frameworks (Axum, Actix)
go.modGo + go modules + frameworks (Gin, Echo)
requirements.txtPython (fallback)
GemfileRuby
composer.jsonPHP

Priority rule: pyproject.toml / Cargo.toml / go.mod take priority over package.json. Only sets values that are actually detected — no hardcoded defaults.


All 12 Tools

Create & Detect

ToolWhat it does
faf_initCreate a starter .faf file with project name, goal, and language
faf_autoAuto-detect stack from manifest files and generate/update .faf
faf_discoverFind .faf files in the project tree

Validate & Score

ToolWhat it does
faf_validateFull Mk4 validation — score, tier, slot counts, errors, warnings
faf_scoreQuick Mk4 score — score, tier, populated/active/total slot counts

Read & Transform

ToolWhat it does
faf_readParse a .faf file into structured data
faf_stringifyConvert parsed FAF data back to clean YAML
faf_contextGet Gemini-optimized context (project + stack + score)

Export & Interop

ToolWhat it does
faf_geminiExport GEMINI.md with YAML frontmatter for Gemini CLI
faf_agentsExport AGENTS.md for OpenAI Codex, Cursor, and other AI tools

Reference

ToolWhat it does
faf_aboutFAF format info — IANA registration, version, ecosystem
faf_modelGet a 100% Trophy-scored example .faf for any of 15 project types

Score and Tier System

Your .faf file is scored on completeness — how many slots are filled with real values.

ScoreTierMeaning
100%TROPHYAI has full context for your project
99%GOLDExceptional
95%SILVERTop tier
85%BRONZEMinimum recommended — AI can build from here
70%GREENSolid foundation
55%YELLOWNeeds improvement
<55%REDMajor gaps — AI will guess
0%WHITEEmpty

Aim for Bronze (85%+). That's where AI stops guessing and starts knowing.


Using with Gemini CLI

> Create a .faf file for my Python FastAPI project
> Auto-detect my project and fill in the stack
> Score my .faf and show what's missing
> Export GEMINI.md for this project
> Show me a 100% example for an MCP server
> What is FAF and how does it work?
> Read my project.faf and summarize the stack
> Validate my .faf and fix the warnings

Architecture

gemini-faf-mcp v2.4.2
├── server.py              → FastMCP MCP server (12 tools, dual-transport, Mk4 scoring)
├── safe_path.py           → path confinement for caller-supplied `path` args
├── main.py                → Cloud Run REST API (GET/POST/PUT)
├── models.py              → 15 project type examples
└── src/gemini_faf_mcp/    → Python SDK (FAFClient, parser)

The MCP server delegates to faf-python-sdk for parsing, validation, and Mk4 scoring. Stack detection in faf_auto is Python-native — no external CLI dependencies.


Testing

pip install -e ".[dev]"
python -m pytest tests/ -v

233 tests passing across 9 WJTTC tiers (137 MCP server + 55 Cloud Function + 41 Mk4 WJTTC championship). Championship-grade test coverage — WJTTC certified.


FAF Ecosystem

One format, every AI platform.

PackagePlatformRegistry
claude-faf-mcpAnthropicnpm + MCP #2759
gemini-faf-mcpGooglePyPI
grok-faf-mcpxAInpm
rust-faf-mcpRustcrates.io
faf-cliUniversalnpm

Python SDK

Use FAF directly in Python without MCP:

from gemini_faf_mcp import FAFClient, parse_faf, validate_faf, find_faf_file

# Parse and validate locally
data = parse_faf("project.faf")
result = validate_faf(data)
print(f"Score: {result['score']}%, Tier: {result['tier']}")

# Find .faf files automatically
faf_path = find_faf_file(".")

# Or use the Cloud Run endpoint
client = FAFClient()
dna = client.get_project_dna()

Cloud Run REST API

Live endpoint for badges, multi-agent context brokering, and voice-to-FAF mutations.

https://faf-source-of-truth-631316210911.us-east1.run.app

Supports agent-optimized responses (Gemini, Claude, Grok, Jules, Codex/Copilot/Cursor) via X-FAF-Agent header. Voice mutations via Gemini Live through PUT endpoint. Auto-deploys via Cloud Build on push to main.


If gemini-faf-mcp has been useful, consider starring the repo — it helps others find it.


Links

License

MIT


Built by @wolfe_jam | wolfejam.dev


Get the CLI

faf-cli — The original AI-Context CLI. A must-have for every builder.

npx faf-cli auto

Anthropic MCP #2759 · IANA Registered: application/vnd.faf+yaml · faf.one · npm

Ver o original no GitHub

Bom para

Relacionados