Access Granted

    Second Brain
    Boilerplate

    Here's the repo. Open it in Claude Code, run the mega prompt, and build your first Second Brain.

    Open the Repo on GitHub

    https://github.com/iwo-szapar/ai-second-brain-boilerplate

    Quick Start (4 Steps)

    1. 1. Clone the repo, or click Use this template / Fork on GitHub, then open the folder in Claude Code
    2. 2. Paste the contents of MEGA_PROMPT_V2.md to let Claude inspect the repo and personalize it for you
    3. 3. Fill in CLAUDE.md and the memory/ files
    4. 4. Run a workflow from skills/ on real work, then add a memory write-back step

    Prefer the terminal?

    clone + open
    git clone https://github.com/iwo-szapar/ai-second-brain-boilerplate.git
cd ai-second-brain-boilerplate
claude  # or open the folder in your AI coding assistant
    Your First Prompt

    One paste to get going

    The fastest start is the adaptive mega prompt at MEGA_PROMPT_V2.md — it inspects the repo, researches your public context, plays back what it understood, asks only what's missing, proposes workflow candidates, and personalizes the boilerplate for you. (MEGA_PROMPT.md is the shorter v1.)

    If you want something shorter to start:

    first prompt
    Load this repo as my AI Second Brain.

Explain what context you found, what is missing, and the first workflow I should customize.

    Or jump straight into a workflow on the sample data:

    run a workflow
    Load my operating context from this repo.

Then read inbox/messy-week-notes.md and run the daily priorities
workflow from skills/daily-priorities.md.
    What's Inside

    The Repo Structure

    ai-second-brain-boilerplate/
    ai-second-brain-boilerplate/
  CLAUDE.md              # who you are + how AI should work with you
  MEGA_PROMPT_V2.md      # adaptive setup prompt (start here)
  MEGA_PROMPT.md         # shorter v1 setup prompt
  README.md
  memory/
    profile.md           # who am I?
    goals.md             # what am I trying to accomplish?
    projects.md          # what's in flight?
    decisions.md         # what did I decide, and why?
    examples.md          # good vs bad output, so AI matches your bar
  skills/
    daily-priorities.md  # turn messy notes into a focused day
    decision-memo.md     # structured decision write-up
  inbox/
    messy-week-notes.md  # sample raw input to run a workflow on
  demo/
    my-second-brain-opener.md
    opening-second-brain-answer.md
    session-checklist.md
  tools/
    allowed-tools.md     # what your AI is allowed to touch

    The core idea: the goal is not a huge knowledge base. It's the smallest set of files that stops you re-explaining your business to AI from scratch. A prompt is useful once. A workflow is useful every week.

    Inside the Folders

    How Each Part Works

    CLAUDE.md is the file your AI reads first in every session — your role, how you want it to work, and the map of the repo. The folders around it hold the durable context so you never re-type it.

    • memory/ — the persistent knowledge: profile, goals, projects, decisions, examples
    • skills/ — repeatable workflows, written once and reused
    • inbox/ — raw, messy input you drop in for the AI to process
    • tools/ — what the AI is allowed to touch

    Day 1 of setup: edit CLAUDE.md with who you are and how AI should work with you.

    Want It Fully Built?

    The boilerplate is the free, do-it-yourself seed. Second Brain 2.0 is the same idea with the memory layer that compounds, 30+ specialist sub-agents, and pre-configured MCPs — done for you.

    Second Brain 2.0

    The complete Claude Code system. 30+ agents, memory that compounds, pre-configured MCPs.

    DIY $197 · Kickstart $597 · DWY $2,497

    See Packages
    MemoryOS

    The memory layer your write-back step writes to. Patterns persist and recall across every future session.

    Free · Standard $199/yr · Pro $349/yr

    Get MemoryOS