MCP server · Coding
OntoPrune MCP server
by vigmarcarlo
Lets your AI read a trimmed-down map of your code instead of whole files, so answers come faster and private code stays private.
OntoPrune is a small helper that sits between your AI assistant and your code files. Instead of handing your AI whole files, it hands over a short summary of what each part does and what it needs from other parts. It is handy for developers working on bigger projects, especially when running a coding AI on their own laptop.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI a new skill or a connection to something else. This one connects your AI to your local code project, so it can look at a trimmed version of your files when you ask it to write or fix code. You add it once, and from then on your AI can call on it during a conversation.
What this MCP server does
You ask your AI to work on a specific function or method in your project. The AI asks OntoPrune for the context it needs. OntoPrune reads your files, keeps only the important shapes (what functions exist, what they take in, what they give back), and skips the private inner workings. It hands that short summary back to the AI, which then writes its answer. There is also a checking step: OntoPrune can look at the AI's answer and tell you if it used anything that was not allowed.
What you can do with it
- Get a short, safe summary of the code around a function you want to change
- Ask your AI to write code using only the allowed parts of your project
- Check whether the AI's answer stuck to the rules or made things up
- Work across Python, Dart, Java, and TypeScript files in the same project
- Run this on your own laptop with a local AI model, without sending code to the cloud
- Use it from the command line if you prefer typing commands over chatting
Try asking your AI
- “Prune the context for the processOrder method in src/services/OrderService.java and show me the stubs”
- “Write the procesar_orden function in services/order_service.py using only the allowed dependencies”
- “Check this generated file against the contract and list any violations”
- “Give me a minimal typed contract for the login function in my TypeScript project”
What it gives back to you
You get back a short block of text showing the function signatures and types your AI is allowed to use. It looks like a list of function names with their inputs and outputs, without the bodies. When you ask it to check code, you get either a short list of problems or a note that everything is fine.
Before you start
What you need
- Python installed on your computer
- The ontoprune package installed with pip
- An AI assistant that supports MCP servers, like Claude Desktop, Cursor, or Gemini CLI
- For local models: Ollama running on your machine
Good to know
It reads your local code files, so make sure you are comfortable with the AI assistant seeing the summaries it produces, even though the private inner parts are left out.
Install it with your AI
Add OntoPrune MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check OntoPrune MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,609 MCP servers here for you.
Who it's for
Developers and engineers who work on multi-file code projects and want faster, safer AI help, especially those running AI models on their own computer.
