Creative Brain

A repo your agent can run to build your personal creative brain.

Clone this repo, give it to Claude Code / Hermes / Cursor / OpenClaw, and tell it:

Use this repo to interview me and build my creative brain.

The agent will interview you about your taste, voice, references, work, and creative instincts, then create a structured markdown brain it can use before writing, designing, brainstorming, or planning for you.

Why this exists

Most people ask AI to create from zero.

That is why the output feels generic.

A creative brain gives your agent context before it starts making things:

• what you make
• what you sound like
• what feels corny to you
• what references shape your taste
• what kind of output feels alive vs dead
• what you are building right now
• what feedback should change next time

The point is not more prompts.

The point is giving your agent a brain to work from.

Who this is for

This is for creatives, builders, creators, designers, editors, writers, operators, and internet people who want AI to understand their taste instead of sounding like a random productivity account.

Install / quickstart

If you just want to use it, read INSTALL.md.

Fastest prompt after cloning:

help me install this and build my creative brain

How to use

Option 1 — with Claude Code / Cursor / OpenClaw

1. Clone or download this repo.
2. Open it in your coding agent.
3. Paste this:

Read README.md and AGENTS.md. Follow the protocol and interview me to build my creative brain inside outputs/creative-brain/.

Option 2 — with Hermes or any chat agent that can edit files

Give the agent this repo and say:

Use the prompts in this repo to interview me. Build my creative brain using the templates. Save the final files under outputs/creative-brain/.

What gets created

Your agent should create:

outputs/creative-brain/
  README.md
  identity.md
  voice.md
  taste.md
  anti-slop.md
  references.md
  content-system.md
  ideas.md
  current-state.md
  feedback-memory.md

The loop

interview → draft brain → feedback → update memory → use brain for future work

The first version will not be perfect.

That is the point.

Every time you approve, reject, or edit output, your agent should update feedback-memory.md so your creative brain gets sharper.

Repo structure

prompts/       agent prompts and run instructions
templates/     blank files the agent copies/fills
examples/      example creative brain fragments
outputs/       where your generated creative brain lives

Important rule

Do not let the agent skip the interview.

If it writes before understanding your taste, it will probably produce slop.