Dify Sync

A CLI tool for bidirectional file synchronization between local files and Dify knowledge bases.

Features

• Bidirectional Sync: Upload local files to Dify or download documents from Dify
• Interactive UI: Terminal-based interface with keyboard navigation
• Progress Tracking: Real-time progress display for uploads and downloads
• File Selection: Multi-select interface for choosing files and documents
• Error Handling: Comprehensive error handling and logging

Installation

npm install -g @takada-at/dify-sync

Configuration

1. Command-line configuration: Run the defy-sync command and configure DIFY settings including the API KEY through the Settings menu.

2. Environment variable configuration:

DIFY_API_URL=https://api.dify.ai/v1/
DIFY_API_KEY=your_api_key_here
DIFY_DATASET_ID=your_dataset_id
LOG_LEVEL=info

Environment Variables

• DIFY_API_URL - Dify API endpoint (must end with /v1/)
• DIFY_API_KEY - Dataset API key from Dify
• DIFY_DATASET_ID - Target dataset ID
• LOG_LEVEL - Logging level (info, debug, error, warn)

Usage

Interactive Mode

Run the application in interactive mode to use the terminal UI:

# Development mode
npm start

# Watch mode
npm run dev

# Built version
npm run build
dify-sync

Command Line Mode

The CLI supports direct upload and download operations:

# Upload files from a directory
dify-sync --upload ./my-files

# Download documents to a directory
dify-sync --download ./downloads

# Force overwrite existing files during download
dify-sync --download ./downloads --force

# Use a specific dataset ID (overrides environment variable)
dify-sync --upload ./files --dataset-id your-dataset-id

# Combine multiple options
dify-sync --upload ./files --dataset-id abc123 --force

CLI Options

• -u, --upload <path> - Upload files from the specified path recursively
• -d, --download <path> - Download files to the specified path
• -f, --force - Force overwrite existing files without confirmation
• --dataset-id <id> - Specify the Dify dataset ID (overrides DIFY_DATASET_ID environment variable)

Note: The --dataset-id option allows you to work with different datasets without modifying your .env file. This is particularly useful when working with multiple Dify datasets or in CI/CD environments.

Commands

Development

• npm start - Run the application directly with tsx (preferred for development)
• npm run dev - Run with file watching (tsx watch)
• npm run build - Build TypeScript to JavaScript in dist/
• npm run typecheck - Run TypeScript type checking

Testing

• npm test - Run tests once
• npm run test:watch - Run tests in watch mode
• npm run test:coverage - Run tests with coverage report

Code Quality

• npm run format - Format code with Prettier
• npm run format:check - Check code formatting

How It Works

The application provides an interactive terminal interface where you can:

1. Upload Files: Select local files to upload to your Dify knowledge base
2. Download Documents: Select documents from Dify to download as local files
3. Track Progress: Monitor upload/download progress in real-time

Navigation

• Use arrow keys to navigate menus
• Press Space to select/deselect files
• Press Enter to confirm selections
• Press Ctrl+C to exit

Architecture

Built with:

• React Ink - Terminal UI framework
• TypeScript - Type-safe development
• Axios - HTTP client for Dify API
• Commander - CLI argument parsing

The application follows a functional programming approach with repository pattern for data access.

API Integration

Integrates with Dify API endpoints:

• Document creation and management
• Dataset operations
• Document segments retrieval

License

MIT