codegraph
Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, and Hermes Agent — fewer tokens, fewer tool calls, 100% local
CodeGraph is an AI sub-agent that provides pre-indexed semantic code intelligence for AI coding assistants by building and maintaining a complete knowledge graph of your codebase. It automatically syncs on file changes, enabling AI agents to retrieve surgical context in a single query instead of performing slow file-by-file exploration. Built with a native Rust kernel, it supports 20+ programming languages, handles framework-aware routing, and runs 100% locally without external API calls.
Key Features
Use Cases
- 01Accelerating AI agent codebase comprehension by providing pre-indexed symbol graphs and call paths in one query instead of many file reads
- 02Answering architecture questions across large repositories without burning tokens on repeated file exploration
- 03Tracing impact radius and dependencies before refactoring to understand which code will be affected by a change
- 04Enabling instant full-text code search across entire projects for AI assistants to locate relevant symbols by name
- 05Maintaining always-current code context for AI coding sessions with automatic incremental sync after every file save
- 06Supporting polyglot projects with unified graphs spanning TypeScript, Python, Rust, Swift, and other languages in one index
Related Agents
View moreunsloth
Local UI to run and train LLMs and diffusion models, including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, FLUX and more.
claude-code-best-practice
from vibe coding to agentic engineering - practice makes claude perfect
system_prompts_leaks
Extracted system prompts from Anthropic - Claude Fable 5, Opus 5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-5.6-Sol, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more. Updated regularly.
impeccable
The design language that makes your AI harness better at design.
codegraph — FAQ
What is CodeGraph?+
CodeGraph is an MCP server that builds and maintains a complete semantic knowledge graph of your codebase, enabling AI coding agents to retrieve surgical context about symbols, dependencies, and call paths in a single query. It runs entirely locally with no external API dependencies.
How do I install CodeGraph and connect it to my AI agent?+
Run the install script (macOS/Linux: curl -fsSL install.sh | sh; Windows: irm install.ps1 | iex) or npm i -g @colbymchenry/codegraph, then run codegraph install to auto-configure your agents, and codegraph init in each project directory to build the code graph.
Which AI coding assistants work with CodeGraph?+
CodeGraph supports Claude Code, Cursor, Codex, Gemini, OpenCode, Hermes Agent, AntiGravity, Kiro, and GitHub Copilot (VS Code, Copilot CLI, JetBrains). The installer detects and configures them automatically.
Does CodeGraph require API keys or external services?+
No. CodeGraph runs 100% locally on your machine with no external API calls, telemetry opt-in only, and all indexing and graph storage kept in your project's .codegraph/ directory.
Is CodeGraph free to use?+
Yes, the open-source CLI and MCP server are free and MIT-licensed. A separate hosted platform for CI/CD impact analysis is in early beta waitlist.
Do I need to manually re-index my code after making changes?+
No. CodeGraph's file watcher automatically syncs the graph on every file change, typically in under one second, so the index is never stale and requires no manual re-run.
How do I install codegraph?+
Open the source repository on GitHub and follow its README. codegraph is a agent — MCP Agents Market links you directly to the official repo.
Is codegraph free?+
codegraph is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.