Tool Calling
also Function Calling
A capability where the LLM outputs a structured request to invoke a defined function or tool, and your code runs it and returns the result for the model to use.
Analogy
Like a chef calling out an order to a line cook: the chef decides what is needed, the cook does the actual work, and the result comes back.
Why it matters
It is how Claude actually performs real actions like sending a message or querying an API instead of just describing them.
In practice
The model returns a call to get_weather({"city":"Manila"}); your code runs it and feeds the temperature back.