AI glossary
What is an AI agent?
An AI agent is a system where a model does more than answer: it decides on steps, uses tools and acts in a loop until it reaches a goal.
A chatbot gets a message and replies. An agent gets a goal ("fix this bug", "put together this week's report") and gets to work: it decides what to do, uses tools like web search, reading files, running code or calling an API, looks at the result and picks the next step.
The difference from an automated workflow is who decides the path. In a workflow the steps are fixed; in an agent the model chooses them as it goes. That makes it more flexible and also less predictable.
That is why it matters to check its work: an agent can say it finished something it did not. Tokencito's agents course is about exactly that, what they really do and what they only say they did.
An example
A coding agent that reads a repository, runs the tests, edits files and runs the tests again until they pass.