Ar9av

Ar9av / obsidian-wiki

Public

Framework for AI agents to build and maintain an Obsidian wiki using Karpathy's LLM Wiki pattern

35
5
100% credibility
Found Apr 06, 2026 at 35 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Framework for AI coding agents to ingest documents, conversations, and project knowledge into an interconnected Markdown wiki viewed in Obsidian.

How It Works

1
📰 Discover the idea

You hear about a clever way to let your AI assistant build and keep a personal knowledge notebook full of your learnings from documents and chats.

2
📥 Bring it home

You download the starter kit and pick a folder for your notes app to hold all your wisdom.

3
🛠️ Easy setup

Run the simple setup helper—it asks for your notes folder and connects everything to your AI assistant automatically.

4
💬 Wake it up

Open the kit in your AI assistant and say 'set up my wiki'—it creates the perfect structure for your knowledge.

5
📚 Feed it knowledge

Use friendly commands like 'ingest this document' or 'update from my chats' to turn files, logs, and conversations into smart, linked notes.

6
🔄 Use from anywhere

In any project, say 'update wiki' to capture new insights or 'ask wiki' to recall what you learned before.

🎉 Smart notebook ready

Your notes app now has a living wiki that grows with you—connected ideas, no repeats, always fresh and easy to search.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 35 to 35 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 obsidian-wiki?

Obsidian-wiki is a Python framework that lets AI agents like Claude Code, Cursor, or Windsurf automatically build and maintain your Obsidian wiki using Andrej Karpathy's LLM wiki pattern. Instead of repeatedly querying LLMs or rebuilding RAG indexes, it distills documents, PDFs, chat histories, and code projects into interconnected markdown pages in your Obsidian vault, tracking changes via deltas to avoid rework. Run slash commands like /wiki-ingest or /wiki-update from any project to keep your knowledge base current and queryable.

Why is it gaining traction?

This agents framework for LLMs stands out with one-command setup.sh that symlinks skills across agents—Claude, Cursor, Windsurf, even Gemini—making it plug-and-play without per-agent tweaks. Developers love the global /wiki-update and /wiki-query commands that work from any repo, plus features like auto cross-linking, linting for broken links, and rebuilds, turning scattered notes into a coherent Obsidian wiki without manual effort.

Who should use this?

AI-assisted coders using Claude Code or Cursor who juggle multiple projects and want to capture architecture decisions, bug patterns, and research into a personal Obsidian wiki. It's ideal for solo devs or teams ingesting chat exports, meeting transcripts, or codebases to avoid forgetting key insights across repos.

Verdict

Early at 35 stars and 1.0% credibility score, but solid docs and multi-agent compatibility make it worth forking for Obsidian wiki fans—test it if you're tired of stale notes. Skip if you need battle-tested production tools.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.