Proto Server

Proto Server

Enables AI assistants to explore and understand Protocol Buffer (.proto) files through fuzzy search, service definitions, and message structure analysis. Integrates with Cursor IDE to provide natural language queries about protobuf schemas and API definitions.

Category
访问服务器

README

Quick Start Guide

Get up and running with MCP Proto Server in under 5 minutes.

Prerequisites

  • Python 3.11 or later
  • pip package manager

Setup Steps

Step 1: Clone the Repository

git clone https://github.com/umuterturk/mcp-proto.git
cd mcp-proto

Step 2: Install Dependencies

⚠️ Required: You must manually install Python dependencies.

pip install -r requirements.txt

This installs:

  • mcp - Model Context Protocol SDK
  • protobuf - Proto parsing support
  • rapidfuzz - Fast fuzzy search
  • watchdog - File watching (optional)

Step 3: Configure Cursor

Add to your Cursor MCP settings file:

macOS: ~/Library/Application Support/Cursor/mcp.json
Windows: %APPDATA%\Cursor\mcp.json

{
  "mcpServers": {
    "proto-server": {
      "command": "python",
      "args": [
        "/absolute/path/to/mcp-proto/mcp_proto_server.py",
        "--root",
        "/absolute/path/to/your/proto/files"
      ]
    }
  }
}

Important:

  • Use absolute paths (not relative)
  • Replace /absolute/path/to/mcp-proto/ with your clone location
  • Replace /absolute/path/to/your/proto/files with your proto directory
  • Or use the included examples/ folder to test

Example for macOS:

{
  "mcpServers": {
    "proto-server": {
      "command": "python",
      "args": [
        "/Users/yourname/mcp-proto/mcp_proto_server.py",
        "--root",
        "/Users/yourname/mcp-proto/examples"
      ]
    }
  }
}

Step 4: Restart Cursor

Close and reopen Cursor. The server will automatically start!

Note: Cursor automatically starts/stops the server. You don't need to run it manually.


Usage

Once configured, ask Cursor questions about your proto files:

  • "What services are available?"
  • "Show me the User message structure"
  • "How do I authenticate?"

The AI will use three MCP tools to explore your protos:

  • search_proto - Fuzzy search across all definitions
  • get_service_definition - Get complete service with all RPCs
  • get_message_definition - Get message with all fields

Troubleshooting

"Module not found" error:

  • Run: pip install -r requirements.txt
  • Check Python version: python --version (need 3.11+)

"No proto files found":

  • Verify the --root path in your config points to a directory with .proto files
  • Test: find /path/to/protos -name "*.proto"

Server not appearing in Cursor:

  • Ensure absolute paths are used in mcp.json
  • Check Cursor's MCP logs for errors
  • Restart Cursor completely

Optional: Testing Before Configuration

Want to verify everything works before configuring Cursor?

Step 3: Test the Installation (Optional)

Run the test suite:

python test_server.py

Expected output:

✓ Indexed 3 proto files
✓ Indexing: PASSED
✓ Search: PASSED
✓ Get Service: PASSED
✓ Get Message: PASSED
✓ Fuzzy Matching: PASSED

Or test the server manually:

# Test with included examples
python mcp_proto_server.py --root examples/

# Test with your own protos
python mcp_proto_server.py --root /path/to/your/protos

Press Ctrl+C to stop the server.


What's Next?

  • USAGE.md - Detailed examples and JSON responses
  • ARCHITECTURE.md - How the system works
  • RECURSIVE_RESOLUTION.md - Efficiency features

Ready to explore your proto files with AI!

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选