MCP Integration
Weproxa includes an embedded Model Context Protocol (MCP) server that exposes proxy tools to external AI clients, enabling AI-assisted HTTP debugging.
What is MCP?
Section titled “What is MCP?”MCP is an open protocol that allows AI models to interact with tools and data sources. Weproxa’s MCP server lets AI assistants access your captured HTTP traffic and proxy tools.
Enabling the MCP Server
Section titled “Enabling the MCP Server”- Open Settings in Weproxa
- Navigate to the MCP section
- Enable the MCP server
- Note the server URL and authentication token displayed
Connecting an AI Client
Section titled “Connecting an AI Client”Configure your MCP-compatible AI client with:
- Server URL — the local HTTP endpoint (e.g.,
http://127.0.0.1:<port>) - Authentication — Bearer token (shown in Weproxa settings)
Available Tools
Section titled “Available Tools”The MCP server exposes tools that let AI assistants:
- List captured requests
- Inspect request/response details
- Filter requests by various criteria
- Interact with proxy features
Security
Section titled “Security”- The MCP server only listens on
127.0.0.1(localhost) - Bearer token authentication is required for all requests
- The server is disabled by default