Persistent memory for coding agents as a single Go binary — SQLite and FTS5 behind MCP, an HTTP API, a CLI and a TUI, with no Node, Python or Docker required.
Burp Suite MCP Server
Summary
PortSwigger's official Burp Suite extension exposing proxy history, Repeater and scanning to AI agents over MCP.
Features
- Runs as a Burp Suite extension, exposing your live Burp session over MCP
- Gives agents access to captured proxy history and existing target scope
- Send modified requests back through Burp from an AI client
- Bundled installer writes the MCP client configuration automatically
- Kotlin implementation, GPL-3.0 licensed, built with Gradle
Installation
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Description
PortSwigger's official MCP server is a Burp Suite extension rather than a standalone process. Once loaded, it opens Burp's own capabilities to any MCP-compatible AI client, so an assistant can work against the traffic and findings already sitting in your Burp session instead of starting a fresh scan from nothing.
That framing is what makes it useful for real web-security work. The proxy history, the requests you have already captured and the target scope you have already defined are the context an agent needs to be worth anything on an engagement — and this server hands it that context directly. An assistant can read through captured traffic, reason about a request, and send modified versions back through Burp, turning the tedious middle of a manual test into something you can direct in prose.
Installation is a build-and-load step: clone the repository, run ./gradlew embedProxyJar, then add the resulting burp-mcp-all.jar through Burp's Extensions tab. The extension ships an installer that writes the MCP client configuration for you. Clients connect through a bundled proxy jar pointed at the Burp MCP server's SSE URL, so Java and the jar command must be on your PATH. It is written in Kotlin and released under GPL-3.0.
The obvious caution applies and is worth stating plainly: this gives an AI model the ability to send crafted HTTP traffic to whatever Burp is pointed at. Use it only against systems you are authorised to test, and keep Burp's target scope tight.
Related MCP Servers
Give a coding agent a live feedback loop on real mobile, TV and desktop apps — inspect, tap, type and capture evidence on iOS, Android and HarmonyOS over MCP or a CLI.
The official W&B MCP server: query experiment runs, Weave LLM traces, artifacts and registries in natural language, and write findings back as a W&B report.
Replicate's official MCP server: search thousands of hosted models, read their schemas, and run predictions on image, video, audio and language models from inside an agent.
