Skip to main content
Graptor turns your entire repository into a structured, vectorized representation giving LLMs the context they need to reason about your code without hallucination.

Install via npm

Get started instantly with a single command.

View on GitHub

Explore the source, contribute, or fork your own version.

​
Why Graptor?

Dharma, a senior engineer at Meta Read full blog on Greptile
I have to touch cross-language repos. One part of the code will be in PHP, and another in Python or C++, and switching between each and seeing how they interact makes it harder to finish the features.
Modern codebases are too large and dynamic to understand manually. Tools like Cursor use LLMs to help but they all need one key thing: A structured understanding of your codebase. Graptor is an open source engine that gives you just that. It reads your entire project, builds a semantic code graph, and then transforms that structure into vector embeddings — the universal language models can understand. For example, Graptor captures:
  • Function declarations
  • Who calls who
The result: a vectorized representation of your codebase a machine readable map that can power any intelligent tooling you build on top of it.

​
What Can You Build With It?

  • LLM-powered code queries – build tools that can answer questions like “Where is this function used?”
  • Dependency visualization – generate structured data to map how modules and functions interact across your repo.
  • AI-native indexing – Embed your code graph into a vector database and run LLM agents on top.
  • Build your own Greptile or Cursor-like tool – powered by real structural understanding, not fuzzy pattern matching.
Graptor is the missing link between static code and intelligent developer tools.

​
Supported Languages

Graptor currently supports JavaScript, TypeScript, out of the box. We’re actively working on adding support for all major languages

​
Learn More

How Graptor works

Understand How Graptor works internally.