Gemini 2.0 API Chat Server

A Go-based API server that provides a chat interface using Google's Gemini 2.0 API with conversation context management and session handling.

Features

• Session Management: Maintains conversation context with a 60-second inactivity timeout
• Base Prompt Integration: Automatically prepends a base prompt to all conversations
• Gemini 2.0 API Integration: Uses Google's Gemini 2.0 Flash Experimental model
• Context Preservation: Keeps conversation history for contextual responses
• Automatic Cleanup: Removes expired sessions to manage memory usage
• Discord Integration: Sends logs and conversation events to Discord webhook

API Endpoints

POST /prompt

Send a prompt and receive a response with conversation context.

Request Body:

{
  "session_id": "unique-session-id",
  "prompt": "Your message here"
}

Response:

{
  "session_id": "unique-session-id",
  "response": "AI response here",
  "error": "error message if any"
}

GET /health

Health check endpoint.

Response:

{
  "status": "healthy"
}

Usage

Starting the Server

# Set your Gemini API key
export GEMINI_API_KEY="your_api_key_here"

# Basic usage (defaults to localhost:8080)
go run main.go config.go

# Custom host and port
go run main.go config.go -h 192.168.1.100 -p 9000

# With custom base prompt file
go run main.go config.go -prompt /path/to/prompt.txt

# All options together
go run main.go config.go -h 0.0.0.0 -p 8080 -prompt custom_prompt.txt

Command Line Options

• -h: Host to bind the server to (default: "localhost")
• -p: Port to bind the server to (default: "8080")
• -prompt: Path to file containing custom base prompt (optional, uses config.go default if not provided)

The server will start on the specified host and port.

Example API Calls

Using curl:

# First message in a session
curl -X POST http://localhost:8080/prompt \
  -H "Content-Type: application/json" \
  -d '{"session_id": "user123", "prompt": "Hello! What can you help me with?"}'

# Follow-up message (maintains context)
curl -X POST http://localhost:8080/prompt \
  -H "Content-Type: application/json" \
  -d '{"session_id": "user123", "prompt": "Can you remember what I just asked?"}'

Testing

Run the test client to see the API in action:

go run client_test.go

Configuration

• Host: Configurable via -h flag (default: "localhost")
• Port: Configurable via -p flag (default: "8080")
• Base Prompt: Configurable via -prompt flag or defined in config.go as fallback
• Session Timeout: 60 seconds of inactivity
• Model: gemini-2.0-flash-exp (Google Gemini 2.0)
• Context Limit: Maximum 20 messages per session
• API Key: Loaded from GEMINI_API_KEY environment variable
• Logging: Timestamped logs with request/response tracking
• Discord Webhook: Integrated for real-time logging and conversation monitoring

Session Management

• Sessions are created automatically when a new session_id is provided
• Each session maintains conversation history
• Sessions expire after 60 seconds of inactivity
• Expired sessions are automatically cleaned up every 30 seconds
• Context is limited to the last 20 messages to manage API costs

Security Notes

• The API key is now loaded from the GEMINI_API_KEY environment variable
• Set the environment variable before running the server: export GEMINI_API_KEY="your_api_key_here"
• To get a Gemini API key, visit: https://makersuite.google.com/app/apikey
• Consider adding authentication and rate limiting
• Validate and sanitize user inputs

Dependencies

• Standard Go libraries only (no external dependencies)
• Requires Go 1.16 or later

Deployment

Railway Deployment

1. Fork/Clone this repository
2. Connect to Railway: Link your GitHub repository to Railway
3. Set Environment Variables:
   • GEMINI_API_KEY: Your Gemini API key
4. Deploy: Railway will automatically build and deploy using the Dockerfile

Docker Deployment

# Build the Docker image
docker build -t chatbot .

# Run with environment variable
docker run -p 8008:8008 -e GEMINI_API_KEY="your_api_key" chatbot

# Or run with custom settings
docker run -p 8008:8008 -e GEMINI_API_KEY="your_api_key" chatbot ./chatbot -h 0.0.0.0 -p 8008 -t 120

Local Development

# Set your Gemini API key
export GEMINI_API_KEY="your_api_key_here"

# Run locally
go run main.go config.go -h 0.0.0.0 -p 8008 -t 120