Right now, it’s being slapped on everything from simple scripts to sophisticated AI workflows. There’s no shared definition, which leaves plenty of room for companies to market basic automation as something much more advanced. Most of today’s agents are powered by large language models (LLMs), which generate probabilistic responses. For the agent model to work, different agents need to cooperate (booking your travel, checking the weather, submitting your expense report) without constant human supervision. It’s meant to be a universal language that lets agents share what they can do and divide up tasks.