Summary

CodeGraphContext indexes local codebases into a Neo4j graph database, providing AI assistants with rich contextual understanding. It targets developers and researchers seeking advanced static analysis, code visualization, and AI-powered code comprehension tools. The primary benefit is enabling deeper insights into code structure and relationships.

How It Works

The project analyzes source code using tree-sitter to construct a knowledge graph within Neo4j. It leverages watchdog for real-time file change detection and graph updates. This approach facilitates sophisticated queries for code relationships (callers, callees, hierarchies) and provides a dynamic, context-aware foundation for AI assistants.

Quick Start & Requirements

Installation is via pip: pip install codegraphcontext. Key dependencies include Neo4j (>=5.15.0), Python, watchdog, typer, and tree-sitter. An interactive setup wizard (cgc setup) guides users through configuring Neo4j (local Docker recommended, Linux binary, or hosted) and integrating with IDEs/CLIs like VS Code, Cursor, Claude, and Gemini. A post-installation PATH fix script may be necessary for the cgc command.

Highlighted Details

• Comprehensive code indexing and relationship analysis (callers, callees, inheritance, call chains).
• Real-time graph updates via file system watching.
• Interactive setup wizard simplifies database and IDE/CLI configuration.
• Broad integration support for popular AI assistants and development environments.
• Features for identifying dead code and calculating cyclomatic complexity.

Maintenance & Community

The project is authored by Shashank Shekhar Singh. Specific community channels (e.g., Discord, Slack) or a public roadmap are not detailed in the provided README snippet.

Licensing & Compatibility

CodeGraphContext is released under the MIT License. This license is permissive, generally allowing for commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

With a version number of 0.1.17, the project appears to be in an early development stage. Users may need to run a provided script to ensure the cgc command is accessible in their environment. The complexity of setting up the Neo4j backend can vary based on the chosen method.