How it works
An agent receives a goal (often a user message), passes it to the LLM along with available tools and conversation history, the LLM decides whether to respond directly or call a tool, the runtime executes any tool call and feeds the result back, and the loop continues until the LLM stops or hits a step limit.
Example
A sales prospecting agent that researches an inbound lead in your CRM, enriches with firmographic data from Clearbit, scores against ICP criteria, drafts a personalised outreach email referencing recent company news, and books a meeting if the lead replies positively.