tronghieu

An AI-powered research assistant for reading, understanding, organizing, and connecting knowledge, implementation of Karpathy's LLM Wiki

24
11
100% credibility
Found May 13, 2026 at 24 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

Lumina-Wiki turns your documents into a structured, AI-queryable personal knowledge base using simple chat commands.

How It Works

1
🔍 Discover Lumina-Wiki

You find this helpful tool and set it up in your project folder with one simple action.

2
📁 Gather your documents

Put your PDFs, notes, and articles into a special folder called raw where your assistant can see them.

3
🧠 Wake up your knowledge assistant

In your AI chat, say a magic phrase like /lumi-ingest, and watch it read everything and build a glowing, organized wiki just for you.

4
💬 Ask smart questions

Chat with your assistant using /lumi-ask to get quick, insightful answers pulled from your personal knowledge base.

5
Add special powers
🔬
Research pack

Find and add new papers automatically.

📖
Reading pack

Break down books chapter by chapter.

💡
Learning pack

Reflect and track how your understanding grows.

Your wiki glows

Sit back and enjoy your second brain – a beautiful, connected web of knowledge that grows with you.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 24 to 24 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 lumina-wiki?

Lumina-wiki is an AI-powered research assistant that builds a structured second brain from your documents. Drop PDFs, notes, or articles into a raw folder, then use chat commands like /lumi-ingest to process them into a linked wiki of summaries, concepts, and people—query it anytime with /lumi-ask. Built in JavaScript with Node.js and Python tools, it installs via npx into any project and integrates with AI chats like Claude or Gemini.

Why is it gaining traction?

It implements Karpathy's LLM Wiki concept cleanly: raw inputs stay separate from AI outputs, avoiding messy chats, with packs adding research discovery, book chapter tracking, or learning reflections. The slash-command workflow feels like an ai powered chatbot github extension, turning tools like Claude Code into a persistent knowledge organizer—no manual note-taking. Obsidian compatibility and arXiv/Semantic Scholar fetching make it a practical ai powered research tool for quick setups.

Who should use this?

Academic researchers surveying papers, book club readers mapping plots and characters, or solo devs curating ai powered projects github notes. Ideal for anyone building a personal wiki around dense sources like research papers or technical docs, especially if you already chat with LLMs daily.

Verdict

Try it if you're into ai powered research assistants—npx install is dead simple, docs are solid, and core workflow shines despite 24 stars and 1.0% credibility score signaling early maturity. Skip for production unless you want to contribute; it's raw potential for tinkerers. (187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.