What is GitNexus (Akon Labs)
GitNexus is an open-source, MCP-native knowledge graph engine that indexes any codebase into a graph of every dependency, call chain, cluster, and execution flow. It provides AI coding agents with precise, structural awareness of the entire codebase, so they never miss code. It is backed by Y Combinator and has gained significant traction on GitHub.
How to use GitNexus (Akon Labs)
- Index your codebase: GitNexus parses every file with Tree-sitter and resolves imports, call chains, field types, and return types across the codebase. This creates a deterministic knowledge graph with no embedding guesswork.
- Cluster the graph: Leiden community detection groups symbols into functional clusters (e.g., auth-core, billing, ingestion) scored by cohesion and modularity, revealing the architecture.
- Analyze impact: Use
detect_changes to map your git diff to affected execution flows and see the blast radius of any change before you commit.
- Query the graph: Agents query hundreds of repos as one graph using MCP tools (query, context, impact, detect_changes, rename, cypher) to get process-level answers: callers, execution flows, blast radius, in a few compact lines.
- Integrate with your agent: Works with Claude Code, Cursor, Antigravity, Codex, Windsurf, and OpenCode.
Features of GitNexus (Akon Labs)
- Deterministic knowledge graph: Resolves every dependency, call chain, field type, and return type across the codebase with 100% determinism and zero guesses.
- Multi-repo support: Groups multiple repositories into one unified graph, enabling cross-repo impact analysis.
- Clustering: Uses Leiden community detection to discover functional clusters and architecture.
- Impact analysis: Traces every downstream caller of a changed function, grouped by depth with confidence scores.
- MCP-native: Provides 7 MCP tools for querying the graph, including
query, context, impact, detect_changes, rename, and cypher.
- Editor integrations: Works with Claude Code, Cursor, Antigravity, Codex, Windsurf, and OpenCode.
- Precomputed analysis: All analyses are precomputed, so the index is reused forever.
- Open source: Available on GitHub with 43.6k stars and 4.9k forks.
Use Cases of GitNexus (Akon Labs)
- AI coding agents: Give agents full codebase structural awareness so they can navigate dependencies, call chains, and execution flows without missing code.
- Impact analysis: Before making a change, see every downstream caller and affected execution flow across multiple repos.
- Architecture discovery: Automatically discover the functional clusters and architecture of a codebase without documentation.
- Cross-repo development: Understand how changes in one service affect dependent repos in a microservices architecture.
- Code review: Use the graph to understand the blast radius of a change and identify potential breakages before they ship.
FAQ
What is GitNexus?
GitNexus is an open-source knowledge graph engine that indexes codebases into a graph of dependencies, call chains, clusters, and execution flows, providing AI agents with precise codebase context.
How does GitNexus work?
It parses files with Tree-sitter, resolves relationships deterministically, clusters symbols using Leiden community detection, and provides MCP tools for querying the graph.
Which editors does GitNexus support?
It works with Claude Code, Cursor, Antigravity, Codex, Windsurf, and OpenCode.
Is GitNexus open source?
Yes, it is open source and available on GitHub with 43.6k stars.
What are the key benefits?
GitNexus improves agent performance (e.g., 68.4% solve rate on DeepSWE vs 37.0% without), reduces token usage and steps, and enables cross-repo impact analysis.
