mcp-architect MCP server

This is a small helper that lets your AI assistant look at a code project on your own computer and describe how it is built. Instead of you pasting a folder list and hoping the AI guesses, it reads the project itself. It is handy for developers and anyone who has to work with a codebase they did not write.

What this MCP server does

You point your AI at a folder on your computer that holds a code project. Your AI passes that folder to this helper, which reads the files locally and works out the languages used, the main entry points, and how the files import each other. It also spots the biggest and most tangled files, and flags any circular dependencies. Your AI then explains all of that back to you in plain words in the chat.

Click to zoom

What you can do with it

• Get a plain overview of a project: languages, frameworks, size, and entry points
• See which files depend on which, and spot circular dependencies
• Find out what would break if you changed a particular file
• List the largest, most complex, or most-changed files
• Ask for a deep dive on any folder or file, including its classes and functions

Try asking your AI

• “Use the architect tools to give me an overview of ~/code/my-app”
• “Show me the dependency graph for this project and any circular dependencies”
• “What breaks if I change the file app/core/registry.py?”
• “Which files in this project are the biggest risks?”

What it gives back to you

You get back plain text answers in the chat: a summary of languages and frameworks, a list of the most depended-on files, any circular dependencies it found, and a ranked list of risky files. For a deep dive it lists classes, functions, and outside libraries used. Nothing is written to your project, it only reads and reports.

Before you start

Who it's for

Developers, tech leads, and anyone who has to understand or review a code project they did not write themselves.