Call Graph LLM Agent

Call graphs. Zero boilerplate. Infinite scale.

• Support any programming languages
• Support any lines of code
• Powered by multi-agents LLM

Architecture

┌─────────────────────────────────────────────────────────────────┐
│                    Input: File or Directory                     │
└────────────────────────────┬────────────────────────────────────┘
                             │
                             ▼
┌─────────────────────────────────────────────────────────────────┐
│                        Doc Agent                                │
│  • Extract function definitions from source code                │
│  • Smart sliding window size for large files                    │
│  • Returns: List of functions with locations                    │
└────────────────────────────┬────────────────────────────────────┘
                             │
                             ▼
┌─────────────────────────────────────────────────────────────────┐
│                     Call Graph Agent                            │
│  • Extract call relationships for each function                 │
│  • Smart sliding window size for function bodies                │
│  • Returns: Edges connecting caller -> callee                   │
└────────────────────────────┬────────────────────────────────────┘
                             │
                             ▼
┌─────────────────────────────────────────────────────────────────┐
│              Output: Call Graph (JSON/Graphviz/Stdout)          │
└─────────────────────────────────────────────────────────────────┘

Usage

Make sure installing dependencies(see Dev) first!

$ ./cga path/to/file.py --llm ollama --llm-config "model=gpt-oss:20b host=http://localhost:11434" --out output/graph.dot -f graphviz
$ ./cga path/to/file.py --llm openai --llm-config "model=gpt-5 api_key=YOUR_KEY" --out callgraph.json -f json

# For example, to generate call graph for example/py-simple/input/main.py with ollama's gpt-oss:20b with graphviz style
$ ./cga callgraph-benchmark/python/project1/main.py --llm ollama --llm-config "model=gpt-oss:20b host=http://<replace with your ollama host>" --out example/py-simple/output/graph.dot -f graphviz

Dev

# if using uv
$ uv sync

# if using uv and need to install dev's dependencies
$ uv sync --extra dev

# if using pip
$ pip install .

# if using pip and need to install dev's dependencies
$ pip install -e ".[dev]"

Contributing

We welcome contributions from the community! 🎉

Ways to Contribute

• Report Bugs: Found a bug? Open an issue with detailed reproduction steps
• Suggest Features: Have ideas for new features? We'd love to hear them!
• Improve Documentation: Help make our docs clearer and more comprehensive
• Submit Code: Fix bugs, add features, or improve performance
• Add Language Support: Help us support more programming languages
• Improve LLM Prompts: Enhance our agent prompts for better accuracy

Development Guidelines

• Write clear, concise commit messages
• Add tests for new features
• Update documentation as needed
• Follow the existing code style
• Keep pull requests focused on a single feature/fix

Need Help?

• Open an issue for questions or discussions
• Check existing issues before creating new ones
• Be respectful and constructive in all interactions

Thank you for helping make Call Graph LLM Agent better! 🚀