Loading the context…
Loading the context…
A software workflow in which an AI model selects actions or uses tools over multiple steps; subagents handle parts of the work.
A task usually needs an objective, allowed tools, a record of progress and a stopping condition. The model receives a result, updates its understanding and decides what to do next. Tools can change real records, so permissions and approval rules must accompany the workflow. Checking the resulting state is more reliable than accepting the model's statement that an action succeeded.
Imagine a newsroom assistant asked to prepare a daily briefing. It finds candidate stories, opens the original reports, compares dates, drafts an original write-up and runs publication checks. If an image lacks usable rights, it searches again or leaves the story pending instead of inventing a credit. A bounded retry limit prevents endless searching. Publication can remain a separate authorized step after the sources and draft are ready.