revibe-mcp
Enables analysis of any GitHub repository to get architecture, file roles, execution flows, system design Q\&A, and structured agent context. Works with MCP-compatible clients like Claude Desktop, Cursor, and Windsurf.
README
revibe-mcp
MCP server for Revibe codebase analysis. Works with Claude Desktop, Cursor, Windsurf, and any MCP-compatible client.
Analyze any GitHub repo — get architecture, file roles, execution flows, system design Q&A, and structured agent context.
Quick Start (Node.js — zero install)
Add to your MCP client config:
{
"mcpServers": {
"revibe": {
"command": "npx",
"args": ["revibe-mcp"]
}
}
}
Then ask your AI assistant to "log in to Revibe" — it will open your browser for a one-time signup.
Quick Start (Python)
pip install revibe-mcp
revibe-mcp-auth login
{
"mcpServers": {
"revibe": {
"command": "revibe-mcp"
}
}
}
Where to add the config
| Client | Config location |
|---|---|
| Claude Desktop | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Claude Code | .mcp.json in your project root |
| Cursor | Settings > MCP Servers > Add |
| Windsurf | MCP config in settings |
Auth
Two options:
Option A: Browser login (recommended)
# Node.js
npx revibe-mcp-auth login
# Python
revibe-mcp-auth login
Opens your browser — sign up or log in, click "Authorize". API key saved automatically to ~/.config/revibe/credentials.json.
Option B: Manual API key
Get a key from app.revibe.codes/settings, then pass it via env:
{
"mcpServers": {
"revibe": {
"command": "npx",
"args": ["revibe-mcp"],
"env": {
"REVIBE_API_KEY": "rk_live_your_key_here"
}
}
}
}
Tools
| Tool | Description |
|---|---|
revibe_login |
Log in via browser — one-time setup |
analyze_repo |
Submit a GitHub repo for analysis (auto-detects from git remote) |
check_status |
Check progress of an analysis job |
get_summary |
Architecture, modules, patterns overview |
get_section |
Drill into architecture, file_roles, system_design_qa, etc. |
get_agent_context |
Full structured JSON optimized for AI agents |
Usage
Once configured, just ask your AI assistant:
- "Analyze this repo with Revibe"
- "What's the architecture of github.com/user/repo?"
- "Get the agent context for this codebase"
The tools handle auto-detection, polling, and structured output automatically.
CLI Commands
# Node.js
npx revibe-mcp-auth login # Log in via browser
npx revibe-mcp-auth status # Show current auth status
npx revibe-mcp-auth logout # Remove saved credentials
# Python
revibe-mcp-auth login
revibe-mcp-auth status
revibe-mcp-auth logout
Claude Code Skill
If you use Claude Code, you can install Revibe as a slash command skill instead of (or in addition to) the MCP server:
# Copy the skill file
mkdir -p ~/.claude/skills/revibe
curl -o ~/.claude/skills/revibe/SKILL.md \
https://raw.githubusercontent.com/selvatuple/revibe-mcp/main/skills/claude-code/SKILL.md
Then use /revibe or /revibe github.com/user/repo inside Claude Code.
The skill file is also available at skills/claude-code/SKILL.md.
Packages & Distribution
| Method | Install |
|---|---|
| npm (zero install) | npx revibe-mcp |
| PyPI | pip install revibe-mcp |
| MCP Registry | registry.modelcontextprotocol.io — search "revibe" |
| Claude Code Skill | Copy skills/claude-code/SKILL.md to ~/.claude/skills/revibe/ |
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。