Overview

CodeGraph is a revolutionary, 100% local code knowledge graph designed to dramatically enhance the capabilities of AI coding assistants. By pre-indexing your entire codebase, CodeGraph provides surgical context directly to agents like Claude Code, Gemini, Cursor, Copilot, and many others. This eliminates the need for slow, token-intensive file crawling and manual dependency analysis, enabling AI to understand and interact with your code with unprecedented speed and accuracy.

Key Features

• Pre-indexed Code Knowledge Graph: A comprehensive graph of all symbols, call edges, and dependencies in your project, built locally.
• Auto-Sync on Code Changes: The graph seamlessly updates in real-time as you modify, add, or delete files, ensuring your AI agent always has the latest, most accurate code context.
• Reduced Token Usage & Tool Calls: By providing direct access to relevant code structures, CodeGraph significantly cuts down on the number of tokens processed and tool invocations required by AI agents, leading to faster responses and lower costs.
• Surgical Context Delivery: Delivers precisely the information needed by the AI, avoiding the overhead of passing entire files or broad search results.
• Broad Agent Compatibility: Supports a wide array of popular AI coding tools, including Claude Code, Gemini, Cursor, Codex, OpenCode, AntiGravity, Kiro, Copilot, and Hermes Agent.
• Extensive Language Support: Works out-of-the-box with a vast array of programming languages, including TypeScript, Python, Go, Rust, Java, C++, Swift, and many more, without per-language configuration.
• Fast, Efficient Kernel: Powered by a high-performance Rust kernel for maximum speed and minimal resource overhead.
• Zero-Knowledge Setup: Easy installation and initialization with a simple CLI.

Typical Use Cases

• Enhanced AI Code Understanding: Enable AI agents to deeply understand complex codebases, their interdependencies, and architectural nuances.
• Faster AI-driven Refactoring and Debugging: Allow AI to pinpoint the exact impact of changes or identify the root cause of bugs with unparalleled precision.
• Improved Code Generation and Review: Provide AI with a complete understanding of existing code patterns and structures for more contextually relevant code generation and insightful reviews.
• Reduced AI Operational Costs: Significantly cut down on token consumption and tool call expenses associated with AI code analysis.
• Real-time Code Context for Agents: Ensure that AI assistants always have up-to-date information, even during active development sessions.
• Streamlined CI/CD and Testing: Lay the groundwork for future features that can precisely identify test coverage gaps or potential regressions based on code changes.