An AI agent with tools is a program where a language model decides when to call functions you provide, such as search or a calculator, and uses the results to finish a task.
In plain words
You describe each tool to the model: its name, what it does and what inputs it takes. When a request needs a tool, the model returns a structured call instead of plain text. Your code runs the function and sends the result back, and the model continues. Repeating this in a loop, checking real results at each step and stopping after a set number of turns, makes an agent.
A quick example
Imagine a weather helper. You give the model a get_weather tool that takes a city. Asked "Do I need a coat in Oslo?", the model calls get_weather with Oslo, your code fetches the forecast, and the model answers using it.
Where you'll meet it
- Assistants that look things up or take actions.
- Coding, research and support agents.
Related
Sources
- DocsTool use with Claude, Anthropic (Claude Platform Docs) · read 21 Oct 2026
- DocsFunction calling, OpenAI API docs · read 21 Oct 2026
- DocsBuilding effective agents, Anthropic · read 21 Oct 2026
Checked by WhatIsAI against the sources above · 21 Oct 2026. Spotted an error? Report it.