🧠 Summary

Added a new example script assistant_weather_tool.py demonstrating how to build a GPT Assistant that can call an external weather API (Open-Meteo) using the Assistants API and function calling.

This recipe helps developers understand how to connect GPT models with real-world APIs through tool integration.

💡 Motivation
The OpenAI Cookbook currently has very few end-to-end examples showing Assistants API + function calling + external APIs.
This contribution makes it easier for developers to:

Define custom tools that call real APIs

Understand tool schema design and integration

Create assistants capable of retrieving live data

📁 Files Added
examples/assistants/assistant_weather_tool.py

✅ Checklist

Code runs successfully locally

Tested with valid OPENAI_API_KEY

Added detailed docstrings and comments

Follows repo style (black, isort)

Uses environment variables (no hardcoded keys)

Provides a practical, original learning example

📸 Example Output
Assistant created. assistant_id: asst_12345
Local tool demo (Open-Meteo):
Fetching weather for New York...
Tool result: {'location': 'New York, United States', 'temperature_c': 22.5, 'windspeed_kmh': 8.6, ...}

👤 Author

Name: Shankar Kumar
GitHub: @Shank312
Email: shankark3121999@gmail.com