Docs

Background tasks

Long-running work keeps going after the reply. How plori backgrounds slow tool calls and posts the result back into the conversation.

A slow command does not hang your conversation. When something takes longer than ten seconds, the agent moves it to the background, answers you, keeps the work running on its computer, and reports back in the same conversation when it finishes.

What happens when a command runs long?

  1. You ask for something that involves a long-running command.
  2. The command starts. If it finishes within ten seconds, everything stays inline.
  3. If it keeps running past ten seconds, the agent moves it to the background, ends its turn, and tells you the job is running.
  4. You can keep talking, or close the conversation entirely.
  5. When the job finishes, the agent picks up the output, summarizes it, and replies in the same thread. If you are watching, the reply streams in live. Otherwise it is waiting the next time you open the conversation.

There is nothing to configure and no polling to write. A result is delivered exactly once: never dropped, never posted twice.

What can run in the background?

Anything the agent can run in its shell: builds, test suites, scrapes, data processing, installs. A single task can run for up to one hour, and each agent can run up to five background tasks at once. Work that should start later rather than run longer belongs in scheduled runs instead.

Where does the output go?

The full output stays on the agent's account disk, so nothing is lost even when output is large. The completion message carries the tail, and you can ask the agent to dig into the rest.