A Go SDK for interacting with the Anthropic Claude API. This SDK allows you to send structured messages and handle responses from Anthropic's conversational AI models seamlessly within your Go applications.
- Easy Integration: Simplifies communication with Anthropic's Claude API using Go.
- Environment Variable Configuration: Securely manage your API keys and settings.
- Detailed Logging: Helps in debugging by logging request and response details.
- Flexible Configuration: Easily adjust parameters like
max_tokens,temperature, and model selection via environment variables.
- Go: Ensure you have Go installed. You can download it from golang.org.
- Anthropic API Key: Obtain your API key from the Anthropic Console.
Clone the repository to your local machine:
git clone https://github.com/aitrailblazer/anthropic-claude-golang-sdk.git
## Configuration
The SDK relies on environment variables for configuration to ensure security and flexibility.
- `ANTHROPIC_API_KEY` (required): Your unique API key for authenticating with Anthropic's API.
- `ANTHROPIC_MODEL` (optional): The model name to use (default is `claude-1.3`).
- `ANTHROPIC_MAX_TOKENS` (optional): Maximum number of tokens to generate in the response (default is 300).
### Setting Environment Variables
On Linux/macOS:
```bash
export ANTHROPIC_API_KEY="your-api-key"
export ANTHROPIC_MODEL="claude-1.3" # Optional
export ANTHROPIC_MAX_TOKENS=300 # OptionalAn example program is provided to demonstrate how to use the SDK to send messages and handle responses.
Navigate to the Examples Directory:
cd anthropic-claude-golang-sdk/examplesRun the Example:
go run example.goExample Output:
2024/09/18 00:28:55 Sending POST request to https://api.anthropic.com/v1/messages with body: {"model":"claude-1.3","messages":[{"role":"user","content":"Hello there."},{"role":"assistant","content":"Hi, I'm Claude. How can I help you?"},{"role":"user","content":"Can you explain LLMs in plain English?"}],"max_tokens":100,"temperature":0.7}
2024/09/18 00:28:57 Received response with status: 200 OK
2024/09/18 00:28:57 Response body: {"id":"msg_01YNjiU5mFbKVy8MNHCP7brd","type":"message","role":"assistant","model":"claude-1.3","content":[{"type":"text","text":"Sure, here's a simple explanation of large language models or LLMs:\n\n• LLMs are AI systems that have been trained on massive amounts of text data. They learn patterns and relationships between words and phrases to understand language.\n\n• The more text data they are trained on, the more they can understand the nuances, complexities and ambiguities of natural language. Some recent LLMs have been trained on hundreds of billions of words.\n\n• LLMs can be use"}],"stop_reason":"max_tokens","stop_sequence":null,"usage":{"input_tokens":41,"output_tokens":100}}
Response: Sure, here's a simple explanation of large language models or LLMs:
• LLMs are AI systems that have been trained on massive amounts of text data. They learn patterns and relationships between words and phrases to understand language.
• The more text data they are trained on, the more they can understand the nuances, complexities and ambiguities of natural language. Some recent LLMs have been trained on hundreds of billions of words.
anthropic-claude-golang-sdk/
│
├── api
│ ├── client.go # Handles API requests, authentication, and logging
│ └── message.go # Contains the SendMessage function to interact with the Messages API
│
├── models
│ └── message.go # Defines the request and response structures
├── examples
│ └── example.go # Demonstrates how to use the SDK
├── go.mod # Go module file
└── go.sum # Go module checksum file
Refer to the official API documentation for detailed information on available endpoints and parameters.
The SDK provides detailed logging to assist with debugging. Logs include request and response details.
Comprehensive error handling is implemented to manage API errors and connection issues effectively.
Contributions are welcome! Please refer to the contributing guidelines for more information.
This project is licensed under the MIT License. See the LICENSE file for details.