ProxyKit MCP server

This is a helper that connects your AI assistant to ProxyKit, a tool that sits on your own computer and watches the web traffic going in and out of your apps. Once connected, you can ask your AI in plain words to look at that traffic, create fake replies, or replay requests. It is handy for developers and testers who debug how apps talk to servers.

What this MCP server does

You type a request in plain words, like asking to see recent traffic or to make an endpoint fail sometimes. Your AI sends that request to this helper program. The helper then talks to ProxyKit on your computer, which is already capturing traffic. ProxyKit does the work, like listing requests or adding a new rule, and sends the result back. Your AI shows you the answer in the chat, as a list, a summary, or a confirmation that something changed.

Click to zoom

What you can do with it

• List recent captured requests and filter them by method, URL, or status
• Search across captured URLs for a word or phrase
• Create a mock rule so an endpoint returns a fake response
• Add latency or errors to a route with a chaos rule
• Replay a whole session or a single request
• Compare two capture sessions to spot differences
• Check findings like privacy or performance issues

Try asking your AI

• “List my latest captured traffic in ProxyKit”
• “Mock the checkout endpoint to fail half the time”
• “Add 800 ms of latency to /api/payments and replay the session”
• “Summarize the last 500 captured requests against my baseline”

What it gives back to you

You get back lists of requests, summaries of findings, or details of a single request with headers and a redacted body. When you change something, like creating a mock rule, you get a confirmation that it was added. Everything shows up right in your chat window, so you can keep working without switching apps.

Before you start

Who it's for

Developers, testers, and anyone who debugs how apps talk to servers and wants to control that from their AI chat.