BurpMCP-Ultra

A Burp Suite Professional extension that runs an MCP server and gives an agent the scanner, proxy and other Burp tools

  • 288 ★ on the repository
  • Last commit
MCP server

Broad access

Why this level

  • Sends real HTTP, WebSocket and fuzzing requests at targets inside the test scope, including the active scanner and Intruder
  • Injection, JWT and access-control checks are offensive techniques, legal only against systems you own or are authorized to test
All reasons and checks

cy-s3c/burpmcp-ultra

Install

Manual install

./gradlew shadowJar

Builds the JAR into build/libs. Needs JDK 17-21. Prebuilt JARs are on the Releases page.

Access and checks

How we review

Broad accessThe tool can change data, run commands or handle secrets.

  • Sends real HTTP, WebSocket and fuzzing requests at targets inside the test scope, including the active scanner and Intruder
  • Injection, JWT and access-control checks are offensive techniques, legal only against systems you own or are authorized to test

This is third-party code. Review the repository files before installing.

What it does

BurpMCP-Ultra is a Kotlin extension for Burp Suite Professional. You load its JAR through the Extensions tab, and it starts an MCP server inside Burp. The server exposes the Burp API to an agent as 150 tools across 37 categories: proxy history, the scanner, Collaborator for out-of-band checks and other areas. On top of the basic actions there are ready-made testing flows. One tool confirms common injection classes through verifiable response signs and returns only confirmed findings. Another sends the same request under different access levels and diffs the responses to find authorization flaws. The extension also generates requests from OpenAPI and Swagger schemas, runs custom scan checks written in BCheck and finds secrets in captured traffic. The server listens on a local address and requires the token from the extension's tab on every request. It rejects foreign Host and Origin headers and keeps test requests inside the scope the operator sets. A web dashboard on a separate port shows the captured event stream in real time.

Who it is for. For pentesters and bug bounty hunters who test their own or authorized systems and want to connect Claude Code or another MCP client to Burp.

Good fit when

  • You want to search Burp's proxy history by regex and have the agent flag suspicious requests without reviewing every item by hand
  • In a test environment you are authorized to touch, you need to diff responses across roles to find broken access control
  • You have an OpenAPI or Swagger schema for an internal API and want real requests generated and sent against it
  • You need a custom passive or active check written as a BCheck that Burp's scanner applies automatically
  • You need a Collaborator payload and want to poll DNS, HTTP and SMTP interactions to find blind vulnerabilities

Not a fit when

  • You do not hold a Burp Suite Professional license: the extension only runs inside it
  • You need to test a system without its owner's written authorization: the tool does not stand in for a legal testing scope
  • You need one network gateway for MCP traffic across several machines: the server runs locally, on one host
  • Port 9876 is already taken by PortSwigger's own MCP Server extension: change the port in settings first

Example request

Connect to BurpMCP-Ultra, pull the last hour of proxy history and show me requests that returned a 500

MCP

Transport
sse
Authentication
API key

Limitations

The MCP server and web dashboard listen on localhost only. The author does not treat that as a trust boundary, so every request needs the token. Destructive operations, such as shutting Burp down, are blocked by an operator flag by default, but that flag can be turned on. The transport is SSE only, and for a stable connection through a proxy the author gives a separate Caddy setup. Building from source needs JDK 17-21, and prebuilt JARs are on the releases page.

How to disable. In Burp Suite, open the Extensions tab, select BurpMCP-Ultra in the list and click Remove. That stops the MCP server and dashboard along with the extension.

FAQ

How is BurpMCP-Ultra different from PortSwigger's own MCP Server?

PortSwigger's official extension ships 12 tools. This one has many more, plus flows such as injection confirmation, JWT attacks and OpenAPI import on top.

How do I connect BurpMCP-Ultra to Claude Code?

Build or download the JAR, load it into Burp via Extensions, the Add button, copy the token from the Server tab and add an sse MCP server with that token in an Authorization header to .mcp.json or ~/.claude.json. If your client cannot set headers, the same token can go straight into the URL, as the Server tab's button shows.

Can I limit what an agent can do through BurpMCP-Ultra?

Yes, the operator sets this in Burp itself, and MCP calls cannot change it. The scope gate runs in off, warn or enforce mode and applies to tools that send live requests. A separate flag allows or blocks destructive operations, and every security-relevant call is logged.

How many tools does BurpMCP-Ultra offer?

150 tools across 37 categories. The server shows the exact count for your version in server.tools.size on the Server tab.

Does BurpMCP-Ultra work without Burp Suite Professional?

No, it's an extension of Burp Suite Professional itself and only runs inside it via the Extensions tab. The free Burp Suite Community Edition does not work.

Discussion

No questions or reviews yet.

MCP server and CLI for reverse engineering: an agent inspects binaries, JavaScript apps, APKs and websites on your machine and returns the evidence

MCP serverBroad access369.2KRepository stars

Nine skills that let an agent run Strix pentests, fix validated findings and add security scanning to CI

SkillBroad access71.5KRepository stars

A skills router for authorized reverse engineering, pentest, CTF and threat analysis: it picks a methodology for the task and checks available tools

SkillBroad access43.1KRepository stars
Editors’ pick

A code security audit skill by Cloudflare: the agent runs recon, coverage-led hunting and independent verification of findings, then produces a structured repor

SkillExtended access42.2KRepository stars

A cybersecurity skill library: forensics, incident response, pentesting, cloud and malware analysis mapped to MITRE ATT&CK and NIST

SkillBroad access34.9KRepository stars
Editors’ pick

NVIDIA's open stack for running OpenClaw, Hermes and LangChain Deep Agents in OpenShell sandboxes with network policy and managed inference

CLIBroad access22.8KRepository stars

README in short

The README starts with a tool count comparison against other extensions and JAR build instructions. Connecting MCP clients, a tool table by category, flows on top of the Burp API and the security model follow.

README badge

Report a problem

Foxx AIBurpMCP-Ultra
Foxx AI

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.