Agent
An AI system that works toward a goal by deciding its own next steps and using tools to take actions.
What makes something an agent is that the model, not fixed code, decides what to do next. Describe any agent with four questions: what you hand it (autonomy), whose permissions it uses (identity), what it can change (permissions), and what starts it (trigger).
Also called
- AI agent
Kinds
- Coding agent An agent whose main job is writing and changing software.
- Personal agent An agent that acts on your behalf, using your permissions.
- Background agent An agent that runs in the background, often in the cloud, while you do other work, and reports back when it is done.
Often confused with
- Assistant A general-purpose AI app that you can talk to, such as the Claude app or ChatGPT.
- Workflow automation A rule-based workflow that runs a process from pre-set rules or pre-written code.
How it connects
- Agent harness turns a model into Agent
- A2A connects agent to Agent
- Eval test Agent
- Guardrail filter inputs and outputs of Agent
- Human-on-the-loop watches and can stop Agent
- Human oversight how people check Agent
- Agent sits on the upper rungs of Levels of autonomy
- Agentic is carried out by an Agent
- Agentic workflow has fixed steps unlike an Agent
- Multi-agent system splits one job across many Agent
- Tool is called by an Agent
- Human-in-the-loop approves each action of Agent
- AI washing often misuses the word Agent