harshkedia177

harshkedia177 / axon

Public

Graph-powered code intelligence engine — indexes codebases into a knowledge graph, exposed via MCP tools for AI agents and a CLI for developers.

claude-code code-analysis code-intelligence dead-code-detection knowledge-graph
421
74
100% credibility
Found Feb 21, 2026 at 20 stars 21x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Axon indexes codebases into a knowledge graph to provide insights like dead code detection, call graphs, execution flows, and impact analysis via a command-line tool and AI agent integration.

How It Works

1
🔍 Discover Axon

You hear about Axon, a helpful tool that maps out how all the pieces in your code connect, making it easy to spot unused parts and understand changes.

2
📦 Get it set up

You add Axon to your computer with a simple download, and it's ready to use right away.

3
🚀 Map your project

You point Axon at your code folder, and it quickly builds a smart map showing functions, classes, calls, and flows in seconds.

4
đź’ˇ Ask smart questions

You search for code pieces, see what calls what, find dead code, or check what breaks if you change something—answers appear instantly.

5
Team up with AI
👤
Explore yourself

Use simple searches to dig into your code map anytime.

🧠
Let AI guide you

Your AI buddy uses the map to explain, refactor, or suggest fixes.

âś… Code feels alive

Now you clean up unused code, spot risky changes, and build confidently—your project is clearer and stronger than ever.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 421 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is axon?

Axon indexes Python, TypeScript, and JavaScript codebases into a local knowledge graph, letting you query structural relationships like calls, types, imports, and execution flows via a CLI or raw Cypher. Run `axon analyze .` to build the graph, then `axon impact UserModel` for blast radius before refactors, or `axon dead-code` to spot unreachable symbols—zero cloud, all offline with watch mode for live updates. It exposes the graph via MCP tools for AI agents in Claude Code or Cursor, so they reason over your architecture, not just text.

Why is it gaining traction?

Unlike grep or LSPs that miss connections, Axon fuses call graphs, git coupling, Leiden communities, and hybrid search (BM25 + vectors) for answers like "files that always change together" or "end-to-end login flow." Devs hook on the AI integration—MCP servers with tools like `axon_context` give agents blast radius and dead code smarts, beating keyword hunts. Local KuzuDB storage keeps it fast and private, no API keys.

Who should use this?

Python/TS backend teams auditing monoliths for dead code or tight coupling. Devs using Cursor/Claude Code who want agents aware of types and flows, not hallucinating imports. Architects diffing branches (`axon diff main..feature`) or tracing processes in FastAPI/Express apps.

Verdict

Early alpha (14 stars, 1.0% credibility) with solid docs and tests, but light adoption means watch for bugs in edge cases. Worth pip install for local graph-powered intel if you're tired of manual greps—pair it with agents for quick wins.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.