← All glossary terms

AI & Web Glossary

AI agent

An AI agent is software that uses an AI model to decide on and carry out a sequence of actions in your tools — reading an email, looking up a record, drafting a reply — rather than just answering a single question.

A chatbot answers; an agent acts. An AI agent is given a goal and a set of tools it's allowed to use (your inbox, your CRM, a calendar) and works through the steps itself: read the new email, find the customer's record, check order status, draft a response, flag anything unusual for a human. The model does the judgment; the surrounding software does the doing.

Agents are powerful exactly because they touch real systems, which means the engineering that matters is the boring part: what the agent is allowed to do, what it must never do, and when it has to stop and ask a person. An agent without those limits is a liability, not an asset.

Practical example: lead routing. An agent reads each inquiry, scores it against your ideal-customer criteria, creates the CRM record, assigns it to the right salesperson, and drafts the first reply for approval. Four tools, one goal, human sign-off where it counts. That's the shape of most agent work we build for small teams.

Where this shows up in practice