MehmetGoekce

MehmetGoekce / llm-wiki

Public

Build Karpathy's LLM Wiki with Claude Code. L1/L2 cache architecture. Logseq + Obsidian support.

ai claude-code developer-tools karpathy knowledge-management
13
0
100% credibility
Found Apr 09, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

Setup tool and templates to create an AI-maintained structured wiki with two knowledge layers, integrated into Logseq or Obsidian for personal knowledge management.

How It Works

1
📰 Discover the Idea

You hear about a clever way to let an AI build and organize a personal wiki that grows with your knowledge, keeping everything linked and up-to-date.

2
📥 Get the Setup Helper

You download a simple helper tool that makes starting your wiki super easy in just minutes.

3
🚀 Run the Easy Setup

You launch the setup, pick your favorite note-taking spot on your computer, and choose categories for your knowledge like projects or learning.

4
Pick Your Note Style
📋
Outline Style

Choose the app that loves bullet points and easy expansions for growing notes.

📄
Flat Pages

Pick the app with simple pages perfect for quick reading and editing.

5
📁 Watch It Build Your Wiki

The tool creates folders, starting pages, and connections automatically, so your wiki is ready to grow right away.

6
🤖 Team Up with Your AI Helper

Link it to your AI chat buddy, who now has special tricks to add info, search, and check your wiki anytime.

7
➕ Feed It New Knowledge

Tell your AI to digest articles, notes, or ideas, and it smartly updates pages with links and details.

🌐 Enjoy Your Living Knowledge Web

Your wiki becomes a connected map of everything you know, always fresh, searchable, and expanding effortlessly.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 13 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 llm-wiki?

llm-wiki is a Shell-based setup script that implements Andrej Karpathy's LLM Wiki concept, turning Claude Code into an automated maintainer for your personal knowledge base. You feed it sources via `/wiki ingest`, and it extracts facts, creates cross-linked pages in Logseq or Obsidian, queries them on demand with `/wiki query`, and runs health checks via `/wiki lint`. It solves the problem of stale, disconnected notes by building a living wiki that compounds knowledge over time, complete with L1 memory for rules and L2 for deep dives.

Why is it gaining traction?

Its 5-minute `./setup.sh` handles schema, git tracking, and tool detection, unlike Karpathy's raw gist which lacks wiring for real tools. The L1/L2 cache keeps critical rules always loaded while pulling project histories only when needed, plus built-in linting catches orphans and leaks. Developers dig the ingest flow that touches 5-15 pages per command, turning raw text into structured, attributed content without manual edits.

Who should use this?

Claude Code users building personal wikis for project histories, workflows, or research notes. Indie hackers tracking business metrics across ingests, or devs maintaining tech stacks like Andrej Karpathy's build GPT from scratch experiments. Logseq/Obsidian fans wanting an LLM to enforce schemas and backlinks without switching apps.

Verdict

Grab it if you're already in the Claude/Logseq ecosystem—solid docs and commands make it instantly usable despite 13 stars and 1.0% credibility score signaling early maturity. Test on a side project first; lacks broad testing but nails the Karpathy LLM wiki vision out of the gate.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.