🦖 🤖 docusaurus-plugin-chat-page

A Docusaurus plugin that adds an AI-powered chat interface to your documentation site. Users can ask questions about your documentation and receive contextually relevant answers powered by OpenAI's GPT models.

Features

• 🤖 AI-powered documentation assistant
• 🔍 Semantic search using embeddings
• 💨 Fast client-side similarity search
• 🏗️ Build-time content processing
• 🔒 Secure (API keys only used at build time)
• 💅 Beautiful UI that matches your Docusaurus theme
• ⚡ Real-time streaming responses
• 📱 Responsive design

How It Works

1. Build Time:

   • Processes your documentation content
   • Splits content into manageable chunks
   • Generates embeddings using OpenAI's API
   • Creates a static JSON file with content and embeddings

2. Runtime:

   • Performs client-side similarity search to find relevant documentation
   • Uses OpenAI's Chat API to generate contextual answers
   • Streams responses in real-time for better UX

Installation

npm install docusaurus-plugin-chat-page
# or
yarn add docusaurus-plugin-chat-page

Configuration

Add the plugin to your docusaurus.config.js:

module.exports = {
  // ...
  plugins: [
    [
      "docusaurus-plugin-chat-page",
      {
        path: "chat", // URL path for the chat page
        openai: {
          apiKey: process.env.OPENAI_API_KEY, // Your OpenAI API key
        },
      },
    ],
  ],
}

Adding to Navigation

To add the chat page to your site's navigation bar, update the themeConfig in your docusaurus.config.js:

module.exports = {
  // ...
  themeConfig: {
    navbar: {
      items: [
        // ... your other navbar items
        {
          to: "/chat", // Make sure this matches your plugin's path configuration
          label: "Chat",
          position: "left",
        },
        // ...
      ],
    },
  },
}

Development Mode

To develop and test the chat interface without incurring OpenAI API costs, you can enable development mode:

module.exports = {
  // ...
  plugins: [
    [
      "docusaurus-plugin-chat-page",
      {
        path: "chat",
        development: {
          mockData: true, // Enable mock data for development
        },
      },
    ],
  ],
}

What Development Mode Does

When mockData: true is set:

• No API key required: The plugin works without an OpenAI API key
• Mock embeddings: Generates deterministic fake embeddings for documentation
• Mock responses: Returns development-mode responses instead of real AI answers
• Visual indicator: Shows a banner indicating development mode is active
• Cost-free: No OpenAI API calls are made

Use Cases

• Local development: Test UI/UX without API costs
• Team collaboration: Developers can work without sharing API keys
• CI/CD: Build and test in pipelines without secrets
• Demos: Create screenshots and demos without real API calls

Important Notes

• Development mode is for testing only - responses are not real AI answers
• A warning banner appears in the UI when mock data is enabled
• Console warnings will indicate when mock services are being used
• Production builds with mockData: true will show a warning

Environment Variables

Create a .env file in your project root:

OPENAI_API_KEY=your-api-key-here

Usage

Once installed and configured, the plugin will:

1. Add a chat page to your documentation site at /chat (or your configured path)
2. Process your documentation during the build phase
3. Enable users to ask questions about your documentation

Users can:

• Ask questions in natural language
• Get AI-generated answers based on your documentation content
• See source references for the answers
• View conversation history

Requirements

• Docusaurus v2 or higher
• Node.js 16 or higher
• OpenAI API key (required for production; optional in development mode)

Security

• OpenAI API key is only used at build time for generating embeddings
• No sensitive data is exposed to the client
• All API calls are made with appropriate security headers

Contributing

Contributions are welcome! Please read our contributing guidelines for details.

License

MIT

Support

If you encounter any issues or have questions, please file an issue on GitHub.