Brainfaq

Provides a Brainfuck debugging MCP server to run and inspect Brainfuck programs with an LLM or debugger agent.

javascript

0

GitHub Stars

javascript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "labarilem-brainfaq-mcp": {
      "command": "npx",
      "args": [
        "brainfaq-mcp"
      ]
    }
  }
}

Brainfaq MCP is a server that enables a Brainfuck debugging session through an MCP client, allowing your preferred AI assistant to inspect and interact with Brainfuck programs in real time for easier debugging and learning.

How to use

You run the MCP server locally using the following command. It starts in stdio mode, allowing you to connect via compatible MCP clients and IDE integrations.

npx brainfaq-mcp

To integrate with Visual Studio Code, create a file at .vscode/mcp.json with this content. This config launches the MCP server when you start the client from VS Code.

{
  "servers": {
    "brainfaq_mcp": {
      "command": "npx",
      "args": ["brainfaq-mcp"]
    }
  }
}

You can also add the MCP server to other IDEs that support LLM agents by configuring them to run the same command. The MCP server will start in stdio mode, allowing your agent to send debugging commands and receive state and outputs.

How to install

Prerequisites you need on your machine to run the MCP server.

node -v
npm -v

Install the project dependencies and prepare the runtime environment. Use the commands below to set up and run the MCP server.

npm i

# When ready to build or run, you can start the MCP server with the standard run command shown above

Additional notes

This MCP server provides full Brainfuck support with memory tape management, loop handling, and input/output integration. It validates bracket matching, detects arithmetic overflow or underflow based on configurable limits, and offers step-by-step execution and state inspection to help you debug programs.

Available tools

load_code

Reset the debugger and load new Brainfuck source code with configurable tape size, min/max cell values, and initial input.

step

Execute a specified number of instructions (default 1) and provide detailed state output.

run

Run the program until completion or until it waits for input, with an optional instruction limit.

add_input

Append characters to the input buffer when the program requests input.

get_state

Return the current interpreter state including memory, pointers, and output, with optional windowing.

read_output

Return the complete output string produced so far.