Proxy AI Inference Server

A simple Express.js server with CORS (Cross-Origin Resource Sharing) enabled for development and API testing.

Features

• ✅ Express.js server with CORS
• ✅ JSON body parsing and error handling
• ✅ Health check endpoint with service status
• ✅ Cohere AI inference endpoint
• ✅ Google Cloud Storage proxy with secure URLs
• ✅ File upload, download, and management
• ✅ API documentation endpoint
• ✅ Environment variable configuration

Setup

1. Install dependencies:

npm install

2. Set up environment variables:

# Create .env file and add your API keys
echo "COHERE_API_KEY=your_cohere_api_key_here" > .env
echo "GOOGLE_CLOUD_BUCKET_NAME=your_bucket_name" >> .env

# Google Cloud Configuration (choose one option):
# Option 1: Using key file
echo "GCP_PROJECT_ID=your_project_id" >> .env
echo "GCP_KEY_FILE=path/to/service-account-key.json" >> .env

# Option 2: Using individual credentials
echo "GCP_PROJECT_ID=your_project_id" >> .env
echo "GCP_CLIENT_EMAIL=your-service-account@project.iam.gserviceaccount.com" >> .env
echo "GCP_PRIVATE_KEY=-----BEGIN PRIVATE KEY-----\n..." >> .env

# Option 3: Legacy variable names (still supported)
echo "GOOGLE_CLOUD_PROJECT_ID=your_project_id" >> .env
echo "GOOGLE_CLOUD_KEY_FILE=path/to/service-account-key.json" >> .env

3. Start the server:

# Development mode (with auto-restart)
npm run dev

# Production mode
npm start

Usage

The server will start on http://localhost:3000 (or the port specified in the PORT environment variable).

Available Endpoints

• GET / - Server status and version info
• GET /api/health - Health check with service status
• GET /api/models - Available AI models information
• GET /api/docs - API documentation
• POST /api/data - Example endpoint that accepts JSON data
• POST /api/cohere/inference - Cohere AI text generation
• POST /api/gcs/upload - Upload file to Google Cloud Storage
• GET /api/gcs/download/:fileName - Get signed URL for file download
• GET /api/gcs/files - List all files in GCS bucket
• DELETE /api/gcs/files/:fileName - Delete file from GCS bucket
• GET /api/gcs/files/:fileName/metadata - Get file metadata

Testing with curl

# Test the root endpoint
curl http://localhost:3000/

# Test the health endpoint
curl http://localhost:3000/api/health

# Test POST endpoint
curl -X POST http://localhost:3000/api/data \
  -H "Content-Type: application/json" \
  -d '{"data": "Hello World"}'

# Test Cohere inference
curl -X POST http://localhost:3000/api/cohere/inference \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Write a short story about a robot", "model": "command", "max_tokens": 100}'

# Get available models
curl http://localhost:3000/api/models

# Get API documentation
curl http://localhost:3000/api/docs

# Test GCS file upload (replace with actual file)
curl -X POST http://localhost:3000/api/gcs/upload \
  -F "file=@/path/to/your/file.jpg"

# List GCS files
curl http://localhost:3000/api/gcs/files

# Get download URL for a file
curl http://localhost:3000/api/gcs/download/1234567890-abc123.jpg

### CORS Configuration

The server is configured to allow all origins for development. For production, you should modify the `corsOptions` in `server.js` to specify allowed origins:

```javascript
const allowedOrigins = ['http://localhost:3000', 'https://yourdomain.com'];

Environment Variables

• PORT - Server port (default: 3000)
• COHERE_API_KEY - Your Cohere API key
• GOOGLE_CLOUD_BUCKET_NAME - Your GCS bucket name

Google Cloud Configuration (choose one option):

Option 1: Using key file

• GCP_PROJECT_ID - Your Google Cloud Project ID
• GCP_KEY_FILE - Path to service account key file

Option 2: Using individual credentials

• GCP_PROJECT_ID - Your Google Cloud Project ID
• GCP_CLIENT_EMAIL - Service account email
• GCP_PRIVATE_KEY - Service account private key (with \n for newlines)

Option 3: Legacy variable names (still supported)

• GOOGLE_CLOUD_PROJECT_ID - Your Google Cloud Project ID

• GOOGLE_CLOUD_KEY_FILE - Path to service account key file

• NODE_ENV - Environment (development/production)

Scripts

• npm start - Start the server
• npm run dev - Start the server with nodemon for development