proxypin-mcp
Enables AI to analyze real-time HTTP(S) traffic captured by ProxyPin, with tools to browse, search, and inspect requests and responses, as well as access saved history sessions.
README
proxypin-mcp
MCP server for ProxyPin — let AI analyze your HTTP(S) traffic in real-time.
How it works
┌─────────────────────────────────────────────────────────────┐
│ ProxyPin App │
│ │
│ MITM Proxy ──▶ EventListener ──▶ WsTrafficServer :12080 │
│ │ ▲ │
│ push │ │ commands │
│ │ │ (history) │
└─────────────────────────────────────────┼───┼──────────────┘
│ │
WS │ │ WS
▼ │
┌─────────────────────────────────────────────────────────────┐
│ proxypin-mcp │
│ │
│ Ring Buffer (real-time) ◀── request / response msgs │
│ sendCommand() / pendingCmds ──▶ cmd_reply (history) │
│ │
│ MCP Tools ──▶ Claude Code / AI │
└─────────────────────────────────────────────────────────────┘
Real-time traffic: ProxyPin pushes every request and response through WebSocket. The MCP server buffers them in a ring buffer (default 1000 entries) and exposes them via list_requests / get_request / search_requests.
History: The MCP sends commands (list_histories, get_history, search_history, get_history_detail) over the same WebSocket. ProxyPin reads the HAR files on its end and returns only the fields needed — summaries for list/search, full HAR only for get_history_detail. This keeps AI context size bounded regardless of how many times history tools are called.
Config sync: ProxyPin sends a config message to the MCP on connect and whenever settings change (e.g. history access toggled), keeping the two sides in sync without polling.
Quick Start
1. Enable WebSocket Push in ProxyPin
Settings → WebSocket Push → Enable (default port: 12080)
2. Configure MCP
Add to your .mcp.json:
{
"mcpServers": {
"proxypin": {
"command": "npx",
"args": ["@k186/proxypin-mcp@latest", "--port", "12080"]
}
}
}
3. Use with AI
AI can now call these tools to analyze your traffic:
Real-time tools
- list_requests — Browse recent captured requests
- get_request — Get full request/response details (headers, body, timing)
- search_requests — Filter by URL keyword, HTTP method, or status code
- get_stats — Overview: connection status, status code distribution, top domains
- clear_buffer — Clear the buffer before a new capture session
History tools (requires History Access enabled in ProxyPin settings)
- list_histories — List all saved history sessions
- get_history_requests — Browse requests in a session (summary only)
- search_history — Search across all sessions by URL/method/status
- get_history_detail — Get full HAR entry for a specific request
Options
| Flag | Env Variable | Default | Description |
|---|---|---|---|
--port, -p |
PROXYPIN_PORT |
12080 |
ProxyPin WebSocket push port |
--buffer-size |
PROXYPIN_BUFFER_SIZE |
1000 |
Max requests to keep in memory |
| — | PROXYPIN_WS_URL |
ws://127.0.0.1:{port} |
Full WebSocket URL (overrides port) |
Examples
// Custom port
{ "args": ["@k186/proxypin-mcp@latest", "--port", "9999"] }
// Larger buffer
{ "args": ["@k186/proxypin-mcp@latest", "--port", "12080", "--buffer-size", "5000"] }
// Via environment variables
{
"args": ["@k186/proxypin-mcp@latest"],
"env": { "PROXYPIN_PORT": "12080", "PROXYPIN_BUFFER_SIZE": "2000" }
}
工作原理
┌─────────────────────────────────────────────────────────────┐
│ ProxyPin 应用 │
│ │
│ MITM 代理 ──▶ EventListener ──▶ WsTrafficServer :12080 │
│ │ ▲ │
│ 推送 │ │ 命令 │
│ │ │(历史记录) │
└────────────────────────────────────────┼───┼───────────────┘
│ │
WS │ │ WS
▼ │
┌─────────────────────────────────────────────────────────────┐
│ proxypin-mcp │
│ │
│ 环形缓冲区(实时) ◀── request / response 消息 │
│ sendCommand() / pendingCmds ──▶ cmd_reply(历史) │
│ │
│ MCP 工具 ──▶ Claude Code / AI │
└─────────────────────────────────────────────────────────────┘
实时流量:ProxyPin 将每条请求和响应通过 WebSocket 推送过来,MCP 缓存到环形缓冲区(默认 1000 条),通过 list_requests / get_request / search_requests 暴露给 AI。
历史记录:MCP 通过同一个 WebSocket 发送命令(list_histories / get_history / search_history / get_history_detail),ProxyPin 在自己这边读取 HAR 文件,只返回所需字段——列表/搜索只返回摘要,完整 HAR 仅在 get_history_detail 时返回。无论调用多少次历史工具,AI 上下文大小都在可控范围内。
配置同步:ProxyPin 在客户端连接时以及设置变更时(如切换历史访问开关)向 MCP 推送 config 消息,两端状态保持同步,无需轮询。
快速开始
1. 在 ProxyPin 中开启 WebSocket 推送
设置 → WebSocket推送 → 启用(默认端口:12080)
2. 配置 MCP
在 .mcp.json 中添加:
{
"mcpServers": {
"proxypin": {
"command": "npx",
"args": ["@k186/proxypin-mcp@latest", "--port", "12080"]
}
}
}
3. 让 AI 分析流量
AI 可以调用以下工具分析你的抓包数据:
实时工具
- list_requests — 浏览最近捕获的请求列表
- get_request — 查看完整的请求/响应详情(请求头、响应体、耗时)
- search_requests — 按 URL 关键字、HTTP 方法或状态码过滤
- get_stats — 总览:连接状态、状态码分布、Top 域名
- clear_buffer — 清空缓冲区,开始新的抓包会话
历史工具(需在 ProxyPin 设置中开启"历史访问")
- list_histories — 列出所有历史会话
- get_history_requests — 浏览某个会话的请求(仅摘要)
- search_history — 跨所有会话按 URL/方法/状态码搜索
- get_history_detail — 获取某条请求的完整 HAR 数据
配置参数
| 参数 | 环境变量 | 默认值 | 说明 |
|---|---|---|---|
--port, -p |
PROXYPIN_PORT |
12080 |
ProxyPin WebSocket 推送端口 |
--buffer-size |
PROXYPIN_BUFFER_SIZE |
1000 |
内存中最大缓存请求数 |
| — | PROXYPIN_WS_URL |
ws://127.0.0.1:{port} |
完整 WebSocket 地址(覆盖 port) |
配置示例
// 自定义端口
{ "args": ["@k186/proxypin-mcp@latest", "--port", "9999"] }
// 增大缓冲区
{ "args": ["@k186/proxypin-mcp@latest", "--port", "12080", "--buffer-size", "5000"] }
// 通过环境变量配置
{
"args": ["@k186/proxypin-mcp@latest"],
"env": { "PROXYPIN_PORT": "12080", "PROXYPIN_BUFFER_SIZE": "2000" }
}
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.