NulightJens

NulightJens / ai-second-brain-skills

Public

Two Claude Code skills for building a Karpathy-style LLM wiki — a compounding AI second brain. Install: git clone https://github.com/NulightJens/ai-second-brain-skills.git && ln -s $(pwd)/ai-second-brain-skills/llm-wiki-setup ~/.claude/skills/llm-wiki-setup && ln -s $(pwd)/ai-second-brain-skills/wiki-self-heal ~/.claude/skills/wiki-self-heal

18
3
100% credibility
Found Apr 13, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

This project provides simple instructions for an AI assistant to transform a personal folder of documents into a living, interlinked knowledge wiki that ingests sources, answers queries, and self-maintains without complex setups.

How It Works

1
📰 Discover AI Second Brain

You hear about a clever way to turn your everyday folder of notes and documents into a smart, growing knowledge base that an AI keeps organized and up-to-date.

2
🔗 Connect to Your AI Helper

You place the ready-made instructions into your AI assistant's toolkit so it can understand how to build and care for your personal knowledge vault.

3
🏠 Create Your Knowledge Vault

You simply tell your AI to set up a new folder for your second brain, answer a couple of easy questions about organization, and watch it create the perfect structure.

4
📁 Add Your Notes and Files

You drop your documents, articles, or clips into the dedicated spot in your vault, ready for the AI to learn from them.

5
🔄 Let AI Build the Wiki

You ask your AI to read the new additions and magically weave their insights into connected pages of knowledge that link together like a personal encyclopedia.

6
🔍 Ask Questions and Explore

You start chatting with your AI about topics, comparisons, or gaps, and it draws smart answers from your ever-improving wiki.

7
🩹 Run Weekly Tune-Ups

Once a week, you tell your AI to check for weak spots, fill in blanks, and make your knowledge base even stronger and more connected.

🧠 Your Second Brain Thrives

Now your folder lives as a powerful, self-growing hub of wisdom that compounds with every new idea, always ready to help you think deeper.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 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 ai-second-brain-skills?

This project ships two Claude Code CLI skills to turn any folder into a Karpathy-style LLM wiki—a compounding knowledge base using plain markdown. Drop PDFs, clips, or notes into a raw directory, run ingest commands like "Ingest the new source I just added to raw/", and the AI builds interlinked wiki pages for queries, while self-heal scans fix gaps weekly. It ends chat-based knowledge loss by persisting synthesized insights between sessions, no databases or embeddings required.

Why is it gaining traction?

Stands out for folder-as-app simplicity versus chunky RAG pipelines: git clone, symlink to ~/.claude/skills/ with $(pwd)/ai-second-brain-skills/llm-wiki-setup, and you're querying a git-tracked wiki in minutes. Developers hook on the transparency—read your own index.md like the AI does—plus easy updates via git pull. Reddit threads on two Claude accounts, two Claude Code subscriptions, and two Claude Pro accounts spotlight its fit for multi-account flows, like two GitHub accounts SSH on the same Mac.

Who should use this?

AI tinkerers building personal research vaults on topics like ML papers. Indie devs tired of re-pasting context for project docs. Writers or analysts synthesizing sources without Obsidian plugins or vector stores, especially those juggling two Claude subscriptions or two GitHub accounts same computer in VSCode.

Verdict

Solid for solo use with crisp docs and Karpathy roots, but 18 stars and 1.0% credibility score signal early maturity—test manually before scheduling. Worth a spin if markdown wikis click; otherwise, stick to full RAG tools.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.