Agent execution

How an agent runs: how output reaches the user, the order in which things happen, what is kept between runs, and how agents build on each other.

• Agent outputAn agent produces output with the output command. Output is streamed to the user as multipart content. Each part is written to the stream the moment the comm…
• Order of executionExecution begins in the agent's main function. A function may call other functions with func or parallel, and an LLM may call functions as tools.
• Error handlingAn agent meets errors from three places: its own expressions, the systems it calls, and the limits the runtime enforces. A command that calls an external sys…
• Conversation stateEach time an agent ends successfully or pauses on an ask, the conversation state is encrypted and stored in Search2o Cloud, or handed to your own code when t…
• Agents building on each otherSearch2o has no orchestrator. An agent runs, answers and ends; it cannot run another agent. What lets agents build on each other is the conversation: every a…