obsidian-second-brain

An evolution of Karpathy's LLM Wiki pattern: a vault that rewrites itself.

Every source updates existing pages instead of just appending new ones. Contradictions reconcile automatically. Your vault compounds while you sleep.

31 commands · auto-synthesis · thinking tools that argue with you · live research from X, the web, and YouTube · 4 scheduled agents · 4 role presets

See it in action · All commands · Install · Presets · Discussions

Read the long-form deep dive: "I rebuilt Karpathy's LLM Wiki. Here's what's missing from the original." Why append-only LLM wikis break at scale, the AI-First Vault Principle, and the three bugs I shipped in the first version.

NEW in v0.5 — Research toolkit: /x-read · /x-pulse · /research · /research-deep · /youtube Pull live X posts, web research with citations, and YouTube transcripts straight into your vault as AI-first notes. Vault-first synthesis knows what you already know and fills only the gaps.

Built by Eugeniu Ghelbur · AI Automation Engineer @ Single Grain building in public · sharing what works

---

The Problem

You use Claude every day. Every session starts from scratch. You re-explain everything. The conversation ends. Everything disappears.

You take notes in Obsidian. Hundreds of files. They just sit there. You make the same decision twice because you forgot you made it six months ago. Ideas rot in daily notes. Nobody connects the dots.

Two powerful tools. Completely disconnected.

---

How this extends Karpathy's LLM Wiki

Karpathy's pattern is brilliant. Drop sources, LLM creates wiki pages, ask questions. This skill takes it further:

---

GitHub에서 전체 내용 보기