ThinkyMiner

ThinkyMiner / codeTree

Public

MCP server with 23 tools for structured code understanding via tree-sitter. 10 languages. 999 tests. One-command install.

code-tree-website.vercel.app ai ast claude code-analysis coding-agents
19
1
100% credibility
Found Mar 11, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

codetree provides AI coding assistants with structured code overviews, skeletons, call graphs, and analysis tools across 10 languages to enable precise interactions without full file reads.

How It Works

1
🔍 Discover codetree

You hear about a helpful tool that lets your AI coding buddy quickly grasp the structure of your entire project without reading every line of code.

2
🗂️ Open your project folder

You go to the folder containing your code project on your computer.

3
🚀 Add the smart code reader

With one easy instruction, you connect the tool to your project so it can scan and understand your code's outline.

4
Choose your coding setup
🤖
Claude Code

Tell Claude to use codetree for your project.

✨
Cursor or VS Code

Add a simple note to your project settings.

💨
Other apps

Follow quick steps for your tool.

5
💻 Start coding with AI

Your AI now sees classes, functions, and connections instantly, asking smart questions instead of guessing.

🎉 Smarter, faster coding

Your AI helper works efficiently, understanding your project's map perfectly and giving precise advice right away.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 19 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 codeTree?

codeTree is a Python MCP server delivering 30 tree-sitter tools for AI agents to query code structure in 10 languages like Python, JS, Go, Rust, and Java. Agents fetch file skeletons, symbols, call graphs, references, or dataflow traces instead of entire files, cutting tokens 25x via one-command uvx install in Cursor, VS Code Copilot, or Claude. No embeddings or config—just point to your repo root for instant structured insights.

Why is it gaining traction?

It enables codetree agent-guided tree search for code generation with large language models, outperforming raw file dumps or grep with precise AST outputs like blast radius, dead code detection, taint paths, and git churn. MCP github copilot vscode users love the zero-setup integration and token efficiency; mcp server ai tools like complexity rankings and change impact beat heavyweight LSPs for quick agent queries.

Who should use this?

Backend devs refactoring large Python/Go repos with Cursor or Claude Desktop, needing call graphs or test finders. Frontend teams in JS/TS using VS Code Copilot for clone detection or variable flows. Any AI coder onboarding unfamiliar codebases via repository maps or hot path analysis.

Verdict

Worth testing for mcp github copilot vscode setups—999 tests and solid docs make it reliable despite 10 stars and 1.0% credibility. Early but potent codetree ai extension; mature enough for prototypes, watch for adoption.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.