.faf - Universal Project Context
AI 与智能体by wolfe-jam
适用于各类 MCP 平台的通用项目上下文方案,可自动检测 CLI,并兼容 50 个 MCP tools。
什么是 .faf - Universal Project Context?
适用于各类 MCP 平台的通用项目上下文方案,可自动检测 CLI,并兼容 50 个 MCP tools。
README
Home: wolfe-jam.github.io/faf-mcp · Hosted MCP endpoint: https://ide.faf.one/mcp/v1 (Streamable HTTP)
The MCP you didn't realise you needed, or wanted but didn't know who to ask, is here. Building on 100,000+ downloads across the FAF ecosystem, we bring you faf-mcp to cure your syncing pain and fuel your chosen AI with optimized context, on-demand.
⭐ A star helps other devs discover faf-mcp — despite the downloads, ~3 of 4 devs check stars first.
What's New in 2.3.0 — The Curated Edition
The tools you actually need, up front. faf-mcp now leads with 15 essential tools by default — a clean, focused surface instead of a wall of options. Need everything? Set FAF_TOOLS=all for the full 29 (every tool stays callable by name either way). And every Core tool's description was rewritten so your AI picks the right one, first time.
Define once. Sync everywhere.
You maintain .cursorrules. Your teammate uses AGENTS.md. Someone on the team just switched to Gemini. Every AI tool wants its own context file — and they all say the same thing in different formats.
faf-mcp is the dedicated MCP server for Cursor, Windsurf, Cline, VS Code, and every non-Claude platform. One .faf file in your repo, synced to every format your team needs.
Context for Cursor & IDE agents: faf-cli (v7.1) authors the files this server syncs — bunx faf export --agents, zero-install and git-native. See FAF-CLI for Cursor & IDE agents 👀.
project.faf
│
┌────────┬───────┴───────┬────────────┐
▼ ▼ ▼ ▼
CLAUDE.md AGENTS.md .cursorrules GEMINI.md
(Claude) (Codex) (Cursor) (Gemini)
Quick Start
Everywhere else:
bunx faf-mcp
Add to your MCP config:
{"mcpServers": {"faf": {"command": "bunx", "args": ["faf-mcp"]}}}
| Platform | Config File |
|---|---|
| Cursor | ~/.cursor/mcp.json |
| Windsurf | ~/.codeium/windsurf/mcp_config.json |
| Cline | Cline MCP settings |
| VS Code | MCP extension config |
| Claude Desktop | Use claude-faf-mcp |
Three Ways to Deploy
| Door | Method | Best For |
|---|---|---|
| Hosted | mcpaas.live | Zero-install, point any MCP client to the URL |
| Self-Deploy | Deploy to Vercel | Your own instance, full control |
| Local | bunx faf-mcp | IDE integration via stdio transport |
Hosted (mcpaas.live)
Point your MCP client to https://ide.faf.one/mcp/v1 — no install, no config, no maintenance. Served from 300+ Cloudflare edges with sub-ms cold starts via 2.7KB Zig-WASM engine.
Self-Deploy (Vercel)
Deploy your own MCP server on Vercel in one click. Once deployed, your server exposes:
/health— Health check/info— Server metadata + tool list/mcp/v1— MCP Streamable HTTP transport
Local (npm)
npx faf-mcp
Interop Tools
| Tool | Platform | Action |
|---|---|---|
faf_agents | OpenAI Codex | Import/export/sync AGENTS.md |
faf_cursor | Cursor IDE | Import/export/sync .cursorrules |
faf_gemini | Google Gemini | Import/export/sync GEMINI.md |
faf_conductor | Conductor | Import/export directory structure |
faf_git | GitHub | Author .faf from any repo URL |
# Sync to all formats at once
faf bi-sync --all
# Generate .faf from any GitHub repo
faf_git { url: "https://github.com/facebook/react" }
Core tier: 15 essential tools shown by default; set FAF_TOOLS=all for the full 29 (every tool stays callable by name either way) · with CLI fallback · 9 test suites · 7 bundled parsers
Cloud Sync
Share your FAF context globally via mcpaas.live:
| Tool | Purpose |
|---|---|
faf_cloud_publish | Upload to cloud, get shareable URL |
faf_cloud_fetch | Pull context from cloud |
faf_cloud_list | List available souls |
faf_cloud_search | Search across souls |
faf_cloud_share | Generate share links |
Example Workflow:
# Upload your project.faf
faf_cloud_publish { soul_name: "my-project" }
→ https://mcpaas.live/souls/my-project
# Anyone can fetch it
faf_cloud_fetch { soul_name: "my-project" }
→ Context merged into local project.faf
Zero-install sharing - Recipients need no MCP setup. Served from 300+ Cloudflare edges with <1ms cold starts via 2.7KB Zig-WASM engine.
Eternal Bi-Sync
Your .faf file and your platform context files stay synchronized in milliseconds.
project.faf ←── 8ms ──→ .cursorrules / AGENTS.md / CLAUDE.md / GEMINI.md
Single source of truth
- Update either side → both stay aligned
--allflag syncs to all four formats at once- Zero manual maintenance
- Works across teams, branches, sessions
AI assistants forget. They drift. Every new session, AI starts guessing again. Bi-sync means context never goes stale.
Tier System: From Blind to Optimized
| Tier | Score | Status |
|---|---|---|
| 🏆 TROPHY | 100% | AI never has to guess |
| ★ GOLD | 99%+ | 1 slot from Trophy |
| ◆ SILVER | 95%+ | Close — keep going |
| ◇ BRONZE | 85%+ | Interim — keep going |
| ● GREEN | 70%+ | Interim — keep going |
| ● YELLOW | 55%+ | AI flipping coins |
| ○ RED | <55% | AI working blind |
| ♡ WHITE | 0% | No context at all |
At 55%, AI is guessing half the time. At 100%, AI is optimized.
use>faf | Prompt Pattern
Start every prompt with "Use FAF" to invoke MCP tools:
Use FAF to initialize my project
Use FAF to score my AI-readiness
Use FAF to sync my context
Use FAF to enhance my project
Works on all platforms — stops web search, forces tool usage.
29 MCP Tools
| Tool | Purpose |
|---|---|
faf_init | Initialize project.faf |
faf_score | Check AI-readiness (0-100%) |
faf_sync | Sync context across platforms |
faf_bi_sync | Bi-directional .faf ↔ CLAUDE.md |
faf_read | Parse and validate FAF files |
faf_write | Create/update FAF with validation |
| Interop Tools | |
faf_agents | Import/export/sync AGENTS.md |
faf_cursor | Import/export/sync .cursorrules |
faf_gemini | Import/export/sync GEMINI.md |
faf_conductor | Import/export directory structure |
faf_git | Author .faf from GitHub repo URL |
| Cloud Tools | |
faf_cloud_publish | Upload to mcpaas.live |
faf_cloud_fetch | Pull from cloud |
faf_cloud_list | List souls |
faf_cloud_search | Search souls |
faf_cloud_share | Generate share links |
Plus CLI fallback (via faf-cli):
faf readme- Extract 6 Ws from README (+25-35% boost)faf human-add- Non-interactive YAML merge (6Ws Builder)faf git- GitHub repo analysis without cloning- And 40+ more commands...
Ecosystem
- claude-faf-mcp — Claude Desktop (33 tools)
- faf-cli — Terminal CLI
- faf-wasm — WASM SDK (<5ms scoring)
- faf.one — Official website
If faf-mcp has been useful, consider starring the repo — it helps others find it.
License
MIT License — Free and open source
Zero drift. Eternal sync. AI optimized. 🏆
"It's so logical if it didn't exist, AI would have built it itself" — Claude
常见问题
.faf - Universal Project Context 是什么?
适用于各类 MCP 平台的通用项目上下文方案,可自动检测 CLI,并兼容 50 个 MCP tools。
相关 Skills
Claude接口
by anthropics
面向接入 Claude API、Anthropic SDK 或 Agent SDK 的开发场景,自动识别项目语言并给出对应示例与默认配置,快速搭建 LLM 应用。
✎ 想把Claude能力接进应用或智能体,用claude-api上手快、兼容Anthropic与Agent SDK,集成路径清晰又省心
RAG架构师
by alirezarezvani
聚焦生产级RAG系统设计与优化,覆盖文档切块、检索链路、索引构建、召回评估等关键环节,适合搭建可扩展、高准确率的知识库问答与检索增强应用。
✎ 面向RAG落地,把知识库、向量检索和生成链路系统串联起来,做架构设计时更清晰,也更少踩坑。
多智能体架构
by alirezarezvani
聚焦多智能体系统架构设计,梳理 Supervisor、Swarm、分层和 Pipeline 等模式,覆盖角色定义、通信协作与性能评估,适合规划稳健可扩展的 AI agent 编排方案。
✎ 帮你系统解决多智能体应用的架构设计与协同编排难题,适合构建复杂 AI 工作流,成熟度高、社区认可也很亮眼。
相关 MCP Server
顺序思维
编辑精选by Anthropic
Sequential Thinking 是让 AI 通过动态思维链解决复杂问题的参考服务器。
✎ 这个服务器展示了如何让 Claude 像人类一样逐步推理,适合开发者学习 MCP 的思维链实现。但注意它只是个参考示例,别指望直接用在生产环境里。
知识图谱记忆
编辑精选by Anthropic
Memory 是一个基于本地知识图谱的持久化记忆系统,让 AI 记住长期上下文。
✎ 帮 AI 和智能体补上“记不住”的短板,用本地知识图谱沉淀长期上下文,连续对话更聪明,数据也更可控。
by deusdata
持久化的代码库知识图谱,可跨会话保留上下文,在 session 重启或上下文压缩后仍能继续使用。
✎ 专治 AI 编程助手“会话失忆”,把代码库沉淀为持久知识图谱,重启或压缩上下文后也能无缝续上开发状态。