A tool call is one specific moment when an AI asks a tool to do something and then reads the result. If the model is the brain and tools are its arms and legs, the tool call is the actual reach. Each one has three parts: the agent chooses the tool, sends the input, and reads the output. Function calling is the structured way a model asks for a specific action; the tool call is the working moment you see in practice.
Imagine an office assistant at a desk. Ask which invoices are still unpaid and they can’t answer from memory. They open the accounting software, run the report, read the result, and tell you what they found. Opening the software and running that report is the tool call.
Why you care
Tool calls are where consequences enter the room. A text answer can be wrong, but a tool call can change a file, spend money, send a message, or expose private information. That’s why permissions, review, and clear instructions matter. A good agent uses the right tool, reads the result carefully, and adjusts; a confused one sprays tool calls everywhere.