Radare2 MCP Server

1# Radare2 MCP Server
2 
3[![ci](https://github.com/radareorg/radare2-mcp/actions/workflows/ci.yml/badge.svg)](https://github.com/radareorg/radare2-mcp/actions/workflows/ci.yml)
4[![radare2](https://img.shields.io/badge/radare2-6.0.4-green)](https://github.com/radareorg/radare2)
5 
6<img width="400" alt="r2mcp logo" src="./r2mcp.png" />
7 
8An MCP server to use **radare2** with AI agents such as OpenCode, Mai, VSCode, Claude, CLION, ...
9 
10## Features
11 
12This implementation provides:
13 
14- 💻 Fully written in C using the native r2 APIs
15- 🧩 Works from the CLI, as an r2 plugin and as an MCP server
16- 🔍 Seamless binary analysis with radare2
17- 🔗 Connect to any local or remote r2/iaito session via r2pipe
18- 🔒 Supports readonly mode, sandbox lock and restrict tools
19- 🔩 Fine grained tools configuration
20- 🔁 Direct stdin/stdout communication model
21- 🛠️ Optional raw access to run r2 commands or r2js scripts
22 
23## Installation
24 
25<img width="400" alt="Screenshot_2025-03-22_at_5 34 47_PM" src="https://github.com/user-attachments/assets/5322c3fc-fc07-4770-96a3-5a6d82d439c2" />
26<img width="400" alt="Screenshot_2025-03-22_at_5 36 17_PM" src="https://github.com/user-attachments/assets/132a1de0-6978-4202-8dce-aa3d60551b9a" />
27 
28### Using r2pm
29 
30The simplest way to install the package is by using `r2pm`:
31 
32```bash
33$ r2pm -Uci r2mcp
34```
35 
36The `r2mcp` executable will be copied into r2pm's bindir in your home directory. However, this binary is not supposed to be executed directly from the shell; it will only work when launched by the MCP service handler of your language model of choice.
37 
38```bash
39$ r2pm -r r2mcp
40```
41 
42That's the common mcpServer JSON configuration file:
43 
44```json
45{
46  "mcpServers": {
47    "radare2": {
48      "command": "r2pm",
49      "args": ["-r", "r2mcp"]
50    }
51  }
52}
53```
54 
55### Using Docker
56 
57Alternatively, you can build the Docker image:
58 
59```bash
60docker build -t r2mcp .
61```
62 
63Update your MCP client configuration file (see below) to use the Docker image to use:
64 
65- `"command": "docker"`
66- `"args": ["run", "--rm", "-i", "-v", "/tmp/data:/data", "r2mcp"]`.
67 
68## Configuration
69 
70### Claude Desktop Integration
71 
72In the Claude Desktop app, press `CMD + ,` to open the Developer settings. Edit the configuration file and restart the client after editing the JSON file as explained below:
73 
741. Locate your Claude Desktop configuration file:
75 
76   - macOS: `~/Library/Application Support/Claude/claude_desktop_config.json`
77   - Windows: `%APPDATA%\Claude\claude_desktop_config.json`
78 
792. Add the following to your configuration file:
80 
81```json
82{
83  "mcpServers": {
84    "radare2": {
85      "command": "r2pm",
86      "args": ["-r", "r2mcp"]
87    }
88  }
89}
90```
91 
92## VS Code Integration
93 
94To use r2mcp with GitHub Copilot Chat in Visual Studio Code by [adding it to your user configuration](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server-to-your-user-configuration) (see other options [here](https://code.visualstudio.com/docs/copilot/chat/mcp-servers#_add-an-mcp-server)):
95 
961. Open the Command Palette with `CMD + Shift + P` (macOS) or `Ctrl + Shift + P` (Windows/Linux).
972. Search for and select `Copilot: Open User Configuration` (typically found in `~/Library/Application Support/Code/User/mcp.json` in macOS).
983. Add the following to your configuration file:
99 
100```json
101{
102  "servers": {
103    "radare2": {
104      "type": "stdio",
105      "command": "r2pm",
106      "args": ["-r", "r2mcp"]
107    }
108  },
109  "inputs": []
110}
111```
112 
113## Zed Integration
114 
115You can use r2mcp with Zed as well by [adding it to your configuration](https://zed.dev/docs/ai/mcp):
116 
1171. Open the command palette: `CMD + Shift + P` (macOS) or `Ctrl + Shift + P` (Windows/Linux). 
1182. Search of `agent: open configuration` or search of `settings`.
1193. Add your server as such:
120 
121```json
122  "context_servers": {
123    "r2-mcp-server": {
124      "source": "custom",
125      "command": "r2pm",
126      "args": ["-r", "r2mcp"],
127      "env": {}
128    }
129  }
130```
131Note: you will need another LLM agent, such as Claude, Gemini or else to be able to use it.
132 
133## For Developers
134 
135### Build from Source
136 
137#### Linux/macOS
138 
139To test the server locally, you can build and install it with make:
140 
141```bash
142make install
143```
144 
145This will compile the server and place the `r2mcp` binary in `/usr/local/bin` on macOS.
146 
147#### Windows
148 
149For Windows, just use meson and ninja like it's done in the CI:
150 
151```cmd
152meson b
153ninja -C b
154```
155