Browse / Second brain / second-brain-starter-kit

second-brain-starter-kit

thejoseplatero

An AI-native second brain you can set up in 10 minutes.

Open thejoseplatero/second-brain-starter-kit → On the boundary of the four rules.

Built for one agent

ships CLAUDE.md and nothing for other agents.

In its own words

Aboutthejoseplatero/second-brain-starter-kit

Flat markdown, one magic CLAUDE.md, one working skill.

README.md

An AI-native second brain you can set up in 10 minutes. Flat markdown files, one magic file that programs how Claude behaves inside your system, and one working thinking skill to prove the point.

No app. No database. No subscription. Text files and context.

The people who get the most out of AI aren't the ones with the best prompts. They're the ones with the best context.

Setup (10 minutes)

1. Get the kit.

2. Make it yours. Open CLAUDE.md and edit the marked sections. It is heavily commented. This is the most important 5 minutes of the setup. The file tells Claude who you are, what this system is, and how to help you. The better this file, the better everything else works.

3. Make it YOUR repo.

4. Add the one habit. Add this to your ~/.zshrc (or ~/.bashrc):

Run brain-sync at the end of every day. That is the whole habit. Everything else is additive.

5. Start using it. Open Claude Code inside the folder and just talk:

  • "Capture this: [any thought]"
  • "Process my inbox"
  • "Six hats on [any decision you're facing]"
  • "What does my brain know about [topic]?"

The included skill: Six Thinking Hats

Say "six hats on [decision]" and Claude runs your problem through Edward de Bono's parallel thinking framework: facts, feelings, risks, upside, alternatives, synthesis. It works on day one with zero personal data. See skills/six-thinking-hats.md.

The repository's own description, then 54% of its README. Read the rest on GitHub ↗

Read from the repository's git tree on 2026-08-17.