Binary Ninja MCP logo

Binary Ninja MCP

by fosdickio

This repository provides a Binary Ninja plugin, MCP server, and bridge for integrating Binary Ninja with LLM clients. It enables seamless, real-time interaction between Binary Ninja and MCP clients for enhanced reverse engineering with AI assistance.

View on GitHub

Last updated: N/A

What is Binary Ninja MCP?

Binary Ninja MCP is a plugin and bridge that allows you to connect Binary Ninja, a reverse engineering platform, to Large Language Model (LLM) clients like Claude Desktop. It exposes Binary Ninja's capabilities through an MCP server, enabling AI-assisted reverse engineering workflows.

How to use Binary Ninja MCP?

  1. Install the Binary Ninja MCP plugin through Binary Ninja's Plugin Manager or manually. 2. (Optional) If using Claude Desktop, configure the Claude Desktop bridge following the provided instructions. 3. Start the MCP server in Binary Ninja. 4. Launch your MCP client (e.g., Claude Desktop) and start prompting it about the open binary.

Key features of Binary Ninja MCP

  • Seamless, real-time integration between Binary Ninja and MCP clients

  • Enhanced reverse engineering workflow with AI assistance

  • Support for Claude Desktop as the primary MCP client

  • Extensible for other MCP client integrations

  • Exposes Binary Ninja capabilities through HTTP endpoints

Use cases of Binary Ninja MCP

  • Generating binary analysis reports

  • Renaming functions with AI suggestions

  • Listing functions, data items, and other program elements

  • Decompiling functions with AI assistance

  • Setting and retrieving comments with AI context

FAQ from Binary Ninja MCP

What is MCP?

MCP stands for Message Channel Protocol. It's a protocol used for communication between different applications.

Which MCP clients are supported?

The primary supported client is Claude Desktop, but the MCP server can be used with any client that implements the MCP protocol.

What versions of Binary Ninja are supported?

The README doesn't explicitly state supported versions, but it requires Binary Ninja to function.

What Python version is required?

Python 3.12 or higher is required.

How do I install the plugin manually?

Copy the repository into the Binary Ninja plugins folder.