Graphify

# Graphify

Converts codebases, documentation, and technical files into a searchable knowledge graph for AI coding assistants. It uses local AST parsing to map relationships between code, docs, and configurations.

Made by graphify-labs on GitHub. These notes are Minda's, written from its README; the project's own files and license are in its repository.

## About

What it is: Graphify turns your project into a knowledge graph you can query. It parses your code, documents, and media files into a structure that you search instead of reading individual files.

Who it suits: This tool helps software developers manage large or complex codebases by making AI coding assistants more efficient.

Where to use it: You can use it with many terminal-based coding assistants and AI platforms, including Claude Code, Cursor, GitHub Copilot, and others.

- Code maps: Generate a visual overview of your project structure.
- Path analysis: Trace connections between files and concepts.
- Conceptual explanation: Get breakdowns of specific methods or classes.

How to set it up: You need Python 3.10 or higher and an installer like uv or pipx.

1. Install the CLI tool using your package manager.
2. Run the installation command to register the skill with your assistant.

```bash
uv tool install graphifyy
graphify install
```

How to use it: You trigger the tool in your AI assistant to build and query your project graph.

| Tool | What it does | |---|---| | /graphify . | Maps your entire project | | graphify explain | Details a specific node or concept | | graphify path | Finds the path between two components | | graphify query | Retrieves a subgraph for a question |

```bash
graphify explain "APIRouter"
graphify path "FastAPI" "ModelField"
```

For newcomers and experts: If you are new to this, treat the graph as a custom map of your project. You can ask this map where to look for details instead of reading every file. For experts, the tool builds a deterministic graph using tree-sitter. It avoids the overhead and hallucinations of LLM-only indexing by using a rigid, traversable structure with confidence scores for every connection.

Good to know:

- Supports over 40 programming languages via tree-sitter.
- Performs local parsing where code stays on your machine.
- License: Apache 2.0.

## On GitHub, 5 November 2026

- Stars: 142,478
- Forks: 13,623
- Watching: 413
- Open issues and pull requests: 1,629
- Language: Python
- License: Apache-2.0
- Topics: ai-agents, antigravity, ast, claude-code, code-analysis, code-search, codex, cursor, developer-tools, gemini, graphify, graphrag
- Trending: #2 repository of the day on GitHub trending (5 November 2026)

## Links

- Code: https://github.com/graphify-labs/graphify
- Website: https://www.graphify.com/
- Releases: https://github.com/graphify-labs/graphify/releases
- Issues: https://github.com/graphify-labs/graphify/issues
- Star history: https://www.star-history.com/#graphify-labs/graphify&Date

## Get the code

```
git clone https://github.com/graphify-labs/graphify.git
```

Then follow the setup in its README. Nothing in this kit runs by itself.

README.md  3 KB