Neurodev Mcp

What is Neurodev MCP?

Neurodev MCP is an MCP server that provides AI-powered Python code analysis, test generation, and test execution through static analysis tools such as pylint, flake8, mypy, bandit, radon, and AST inspections, plus auto-formatting with black and autopep8.

How to use Neurodev MCP?

Install via git clone, create a virtual environment, and run pip install -e .. Configure an MCP client (e.g., Claude Desktop or Cline) with the command to launch neurodev_mcp.server. The server supports STDIO (default) and SSE transports. Use natural language prompts like “Review this Python code for issues” to invoke its tools.

Key features of Neurodev MCP

• Comprehensive code review with six static analyzers
• Intelligent pytest test generation via AST analysis
• Isolated test execution with coverage and timeout
• Auto-formatting to PEP8 with black and autopep8

Use cases of Neurodev MCP

• Review Python code for style, security, and complexity issues
• Automatically generate unit tests for functions and classes
• Run tests and get coverage reports in a development workflow
• Format messy code to PEP8 standards quickly

FAQ from Neurodev MCP

What Python version is required?

Python 3.8 or higher is required.

What dependencies does it need?

The server requires packages like pylint, flake8, mypy, bandit, radon, black, autopep8, pytest, pytest-cov, and pytest-timeout (all listed with version constraints in the README).

How do I configure it for Claude Desktop?

Edit ~/Library/Application Support/Claude/claude_desktop_config.json with an absolute path to the virtual environment’s Python and the module arguments.

What transports are available?

The server supports STDIO (default, for local MCP clients) and SSE (Server-Sent Events, for web-based integrations on port 8000 by default).

Can I use it as a library?

Yes, you can import CodeAnalyzer, TestGenerator, and TestExecutor from neurodev_mcp in your own Python async code.