Write Once, Deliver Personally
MarkdownFlow extends standard Markdown with AI to create personalized, interactive pages. Instead of writing static content for readers, you write prompts for AI agents that transform your instructions into personalized content adapting to each user's choices, preferences, and context.
- 📝 Fully Compatible with Markdown - All standard Markdown syntax works unchanged
- 🤖 AI-Powered Personalization - Write prompts for AI, not content for humans
- 🎯 Simple Extensions - Interactive elements, preserved content, and memory
- 🌍 Multi-Language Support - Built with international audiences in mind
| Resource | Description | Link |
|---|---|---|
| 🌐 Official Website | Project homepage and overview | markdownflow.ai |
| 🎮 Playground | Try MarkdownFlow interactively | play.mdflow.run |
| 📚 Documentation | Complete guides and references | markdownflow.ai/docs |
| Project | Description | Language | License |
|---|---|---|---|
| markdown-flow-ui | A React component library to render interactive MarkdownFlow documents with built-in UI elements. | React/TypeScript | MIT |
| remark-flow | A remark plugin to parse and process MarkdownFlow syntax in React applications. | TypeScript | MIT |
| markdown-it-flow | A markdown-it plugin to parse and render MarkdownFlow syntax. (Coming soon) | TypeScript | MIT |
| Project | Description | Language | License |
|---|---|---|---|
| markdown-flow-agent-py | A Python library to transform MarkdownFlow documents into personalized content. | Python | MIT |
| markdown-flow-agent-go | A Go library to transform MarkdownFlow documents into personalized content. (Coming soon) | Go | MIT |
| Project | Description | License |
|---|---|---|
| markdown-flow | The main repository containing homepage, documentation, and interactive playground. | CC |
MarkdownFlow adds interaction, preserved content, and memory to standard Markdown:
Buttons and input fields collect user choices. Within one document, later prompts can refer naturally to earlier answers.
Ask the user about their experience level:
?[Beginner | Intermediate | Expert]
Generate a recommendation based on the experience level the user just selected.
Collect the user's name:
?[...Enter your name here]Text that maintains its exact meaning during AI processing.
Introduce our company values, preserving the exact motto:
===Innovation Through Simplicity===Memory stores questions and answers. Add a question at the start of an interaction to save its answer, then use the same question in later prompts:
Ask the user about their learning goal:
?[{{What would you like to learn?}} Python | JavaScript | ...Another topic]
Design a learning plan based on the answer to {{What would you like to learn?}}.The AI uses the remembered answer as context. The host platform determines where it is saved and which documents can access it. See the Memory specification.
- 📚 Education - Personalized learning materials adapting to student levels
- ✍️ Content Creation - Articles adjusting terminology based on reader expertise
- 📰 News Media - Stories expanding from different angles based on interests
- 🎮 Interactive Storytelling - Narratives where choices influence plot direction
- 📖 Read Core Concepts - Understand fundamental principles
- ⚡ 5-Minute Quick Start - Build your first interactive document
- 🎮 Try the Playground - Practice with live examples
- 🚀 Integration Guide - Add to your project
This project is licensed under Creative Commons. Individual components have their own licenses (MIT for most libraries).
Created by AI Shifu, LLC • © 2025
Core Principle: Write prompts for AI, not content for humans