lucasastorian

lucasastorian / llmwiki

Public

Open Source Implementation of Karpathy's LLM Wiki. Upload documents, connect your Claude account via MCP, and have it write your wiki !

llmwiki.app agents ai-agents claude karpathy knowledge-base
16
5
100% credibility
Found Apr 06, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

LLM Wiki is a free open-source app where you upload documents and let Claude AI build and maintain a structured wiki with summaries, cross-references, diagrams, and citations.

How It Works

1
👋 Discover LLM Wiki

You find this free tool online and sign up with your email or Google account to get started.

2
📚 Create your wiki

Name your personal knowledge space where all your notes and insights will live.

3
📤 Upload your sources

Drag in PDFs, articles, notes, or office files — they become your unchangeable foundation.

4
🤖 Connect Claude

Copy a simple link from settings and paste it into Claude.ai so your AI helper can access everything.

5
✨ Let Claude build

Chat with Claude — it reads your files, creates summaries, links ideas, draws diagrams, and keeps everything organized.

6
🔍 Explore and grow

Browse your wiki pages, ask questions, add more files — it gets smarter with every update.

🚀 Compounding knowledge

Your wiki stays fresh and connected automatically, turning raw info into a powerful, living reference.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 16 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 llmwiki?

Llmwiki is an open-source Andrej Karpathy implementation of the LLM wiki concept, letting you upload documents like PDFs and notes, then connect your Claude account via MCP to automatically generate and maintain a structured wiki. It handles raw sources in one layer, compiles markdown pages with citations and diagrams in another, and exposes MCP tools for search, read, write, and delete operations. Built with TypeScript frontend via Next.js and Python FastAPI backend, it runs on Supabase Postgres and S3 storage, with a live demo at llmwiki.app.

Why is it gaining traction?

This MCP implementation github stands out by turning Claude into a wiki editor that compounds knowledge across ingests—no re-deriving chunks every query. Developers dig the zero-setup hosted version plus self-hosting via Docker Compose, plus OCR for PDFs and office docs, mermaid diagrams, and citation footnotes that render as UI badges. Early buzz comes from Karpathy's llm wiki github spec, making it a ready-to-run alternative to custom RAG setups.

Who should use this?

Researchers synthesizing papers, devs maintaining project wikis from specs and notes, or anyone curating domain knowledge like blockchain implementation github or tcp implementation github explorations. Ideal for solo users or small teams needing Claude to handle wiki housekeeping, cross-references, and lints without manual toil.

Verdict

Grab it if you want Karpathy's vision now—solid docs and demo make the 16 stars and 1.0% credibility score forgivable for an early Apache 2.0 project. Self-host for production, but watch for quota scaling as it matures.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.