Skip to content

billbuilds/mindbridge

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MindBridge

Bridge the gap between AI output and human understanding.

AI generates information at superhuman speed and density. Human brains are biological — they can only absorb information through specific cognitive patterns. MindBridge solves this mismatch.

The Problem

When AI explains something complex, it often dumps a wall of text that's technically correct but cognitively overwhelming. The bottleneck is not the AI — it's the human brain.

The Solution

MindBridge is a collection of markdown files that teach AI agents how to communicate in ways that align with how human brains actually learn. It's built on 4 research-backed learning science frameworks:

Framework Researcher Core Idea
Cognitive Load Theory John Sweller Working memory holds ~4 chunks. Don't overload it.
Feynman Technique Richard Feynman If you can't explain it simply, you don't understand it.
Multimedia Learning Richard Mayer Words + visuals > words alone (with rules).
Progressive Disclosure Nielsen Norman Group Show essentials first, details on demand.

How It Works

MindBridge is zero-code, zero-dependency, zero-install. It's just markdown files that any AI agent can read.

You + AI Agent
      │
      │  Feed these markdown files to your agent:
      │
      ├── SPEC.md              → How to structure content for human brains
      ├── domains/general/     → Universal learning methods
      ├── templates/default.md → Output style guide
      └── skills.md            → Commands: /learn, /html, /md

Quick Start

  1. Copy SPEC.md and skills.md into your AI agent's context
  2. Use the commands:
Command What it does
/learn AI teaches you using the best format for the content
/html Generate a rich HTML preview and open in browser
/md Generate a Markdown document

Works with any AI agent — GitHub Copilot CLI, Claude Code, Cursor, ChatGPT, or anything that reads text.

Project Structure

mindbridge/
├── README.md                    # You are here
├── SPEC.md                      # Core specification (4 frameworks)
├── skills.md                    # Command router (/learn, /html, /md)
├── domains/
│   ├── general/METHOD.md        # Universal learning methods
│   └── _template/METHOD.md      # Template for contributors
├── templates/
│   ├── default.md               # Default output style
│   └── _template.md             # Template for contributors
└── examples/                    # Before/after demos

Contributing

MindBridge is 100% markdown. Anyone can contribute — no coding required.

Add a Learning Domain

  1. Copy domains/_template/METHOD.md
  2. Fill in the research basis and learning principles for your domain
  3. Submit a PR

Add an Output Template

  1. Copy templates/_template.md
  2. Describe the visual style and layout rules
  3. Submit a PR

Philosophy

  • Human brain is the bottleneck, not AI. Design for the reader.
  • Research-backed, not opinion-based. Every principle cites a framework.
  • Zero dependencies. Just markdown. Any AI can read it.
  • Community-driven domains. Experts contribute how their field is best learned.

License

MIT

About

Bridge the gap between AI output and human understanding. Zero-code markdown files that teach AI agents to communicate the way human brains learn.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors