duysolo

duysolo / codebaxing

Public

Local codebase indexing and semantic code search via MCP

21
3
100% credibility
Found Mar 23, 2026 at 21 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Codebaxing indexes codebases to enable natural language semantic search within AI-powered code editors.

How It Works

1
📖 Discover smart code search

You hear about Codebaxing, a helpful tool that lets you find code using simple everyday words instead of exact matches.

2
🗄️ Start the knowledge keeper

You launch a quiet background helper that will store and remember your project's details safely on your computer.

3
📁 Feed it your project

You point it to your project's folder, and it quietly studies every file to understand what your code does (just once).

4
🔗 Link to your AI buddy

You connect it to your favorite AI coding app like Claude or Cursor so they can use this new smarts.

5
🔄 Refresh and go

You restart your coding app, and now everything feels connected and ready to help.

6
🔍 Ask in plain words

You type questions like 'find the login steps' and see perfect matches with file locations pop up instantly.

🎉 Code mastery unlocked

Your AI now knows your whole project inside out, making you super fast at finding and fixing things.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 21 to 21 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 codebaxing?

Codebaxing indexes your local codebase for semantic code search, letting you query with natural language like "find auth logic" to surface relevant functions or classes across files. It parses code with Tree-sitter, generates embeddings via ONNX models, and stores everything in a local ChromaDB instance for fast retrieval. Built in TypeScript as an npm package, it offers CLI tools for indexing, searching, and stats, plus an MCP server that plugs into AI editors for local codebase awareness and RAG.

Why is it gaining traction?

It runs entirely locally—no cloud dependencies—making it a solid local GitHub Copilot alternative or Ollama local codebase companion, with support for 30+ languages and parallel indexing for large repos. The MCP integration means instant semantic search in tools like Cursor or Claude Desktop, plus memory tools to store project notes. Setup is dead simple: Docker for ChromaDB, npx for indexing, and auto-config for editors.

Who should use this?

Backend devs hunting database connection logic in sprawling monorepos, frontend teams querying component patterns without grep, or AI agent builders needing local codebase RAG for Postman-like testing or local GitHub Actions runners. Ideal for anyone with a local GitHub repository wanting semantic search over exact-match tools.

Verdict

Early days with 21 stars and 1.0% credibility score—test coverage is basic, but docs are thorough with Vietnamese translation and clear CLI examples. Worth a spin for local LLM codebase indexing if you use MCP editors; skip if you need battle-tested scale.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.