AI agent glossary
Tool use (function calling)
The ability of a language model to request that your code run a specific function, like a search or database query, and then use the result.
Tool use, also called function calling, is what lets a model act. You describe each tool with a name, a description, and a schema for its inputs. When the model decides a tool would help, it replies with a structured request naming the tool and its arguments.
The model never runs anything itself. Your code receives the request, executes the function, and sends the result back. The model then continues, perhaps calling another tool or giving its final answer. Repeating this exchange is the agent loop.
Tool design has a big effect on agent quality. Clear names and descriptions, focused purposes, helpful error messages, and concise outputs make tools far easier for a model to use correctly.
Given a get_order_status tool, a model asked 'where is my order 4521?' responds with a call to that tool with order_id 4521, then turns the returned status into a friendly reply.
Learn more in the course
Related terms
Go from definitions to doing
Learn to use and build AI agents in 65 practical lessons.