SherwinQ

基于 [Andrej Karpathy]提出的 [LLM Wiki 模式]构建的 Agent Skill,通过四阶段流水线将碎片化信息转化为结构化、可检索、持续增长的个人知识库。

16
3
69% 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
Python
AI Summary

A starter template for setting up an Obsidian vault to build a personal wiki on Andrej Karpathy's AI and machine learning content, with tools to parse PDFs and images into markdown notes using an external service.

How It Works

1
🔍 Discover the Karpathy Wiki

You find this helpful starter kit on GitHub for building your own personal collection of notes on Andrej Karpathy's ideas about AI and learning.

2
📁 Set up your wiki folder

Run a simple setup to create organized folders for your raw files, notes, and logs, all ready for your note-taking app.

3
🔗 Connect a document reader

Add details for a service that can read and understand PDFs and pictures, so you can easily turn them into notes.

4
📖 Open in your note app

Launch your favorite note-taking app like Obsidian and open the new wiki folder as your workspace.

5
Turn files into notes

Pick a PDF or picture, and watch it transform into clean, editable text pages with images saved alongside.

6
📂 Organize your knowledge

Sort the new notes into topic folders, update your dashboard and indexes to see everything at a glance.

🎉 Your wiki is alive

Now you have a beautiful, searchable personal wiki full of Karpathy's wisdom, ready for daily learning and 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 karpathy-wiki?

This Python project sets up a personal knowledge base inspired by Andrej Karpathy's LLM wiki mode from his blog and Twitter posts, turning scattered notes, PDFs, and images into a structured, searchable Obsidian vault. Run a bash init script to create folders for raw inputs, wiki pages, query outputs, and logs, complete with .env for API keys and gitignore. It includes a CLI tool to parse PDFs or images into Markdown via a layout API, extracting text, tables, and images for easy ingestion.

Why is it gaining traction?

Fans of Karpathy's GitHub repos like nanoGPT and micrograd appreciate this wiki karpathy agent starter that automates media-to-wiki conversion, skipping manual OCR hassles. The Obsidian integration and four-stage pipeline for growing knowledge bases stand out over basic note apps, letting you query fragmented info like Karpathy's LLM council demos on YouTube. At 16 stars, it's niche but hooks developers vibe coding their own research wikis.

Who should use this?

Obsidian power users drowning in research PDFs or screenshots from Karpathy's transformers GitHub and autoresearch projects. AI researchers or indie hackers building agent skills for personal KB, especially those following his net worth of ideas on X. Not for teams needing hosted solutions.

Verdict

Worth forking if you're into Karpathy obsidian workflows and have a layout API key—solid starter despite 0.7% credibility score and low maturity from 16 stars. Add tests and full ingest CLI to make it production-ready; great for solo tinkering now.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.