also function calling, tool use
The model returns a structured request to run a named function with arguments; your code runs it and sends the result back. The model never executes anything itself, so permissions, validation and approval live in the code that runs the tool.
You have done this if
The assistant opened a ticket through your API after the model returned a create_ticket call with the fields filled in, and your code checked them first.
Say it in a review
Every tool call is checked against its schema and the caller's permissions before it runs, and anything with side effects waits for approval.
On the AI Application map Tools, Agent
Read Tool access is where AI becomes operational risk · Structured output turns model text into something software can trust