Reviewable agent work
Turn agent output into tracked work.
Tasks make agent work visible, reviewable, and resumable. Use them when an answer needs ownership, a status, acceptance criteria, or a human review loop.
Open Workspaces, choose a workspace, and select Tasks. For the exact filters, statuses, and task-detail behavior, see the Tasks Product Specification.
If task behavior changed after it was created, Agent History helps an administrator check whether the assigned agent's configuration changed.
When To Create A Task
Needs review
A result should be checked before it is sent, published, or used outside the conversation.
Needs an owner
Someone needs to own follow-up work, make a decision, or continue the next step later.
Needs continuity
The agent should resume later with the same context instead of starting from an empty chat.
Needs investigation
A workflow, tool call, or conversation produced a concrete action item or failure to inspect.
Statuses
The task board uses four operating states:
Work is captured but not started.
The owner or agent is actively working.
The output needs a human or teammate check.
The work is complete and no more action is expected.
Review tasks on the workspace's Kanban board. Use search and the Agent filter when the collection grows. Open a card to inspect its prompt and status; with edit access, select Save after changes or drag a card between status columns.
A Good Task Has
For this example, use an agent with Task Management enabled and assigned to the Reporting workspace; naming a different workspace in the prompt does not change where the task is created.
Create a task in the Reporting workspace for the Reporting agent. Goal: compare Q2 feedback themes with last quarter. Output: table of top issues, count, source examples, and recommended next action. Status: Review.
Recover Missing Context
If a task result looks detached from the original request, open the source conversation or task conversation history. If the source is missing, add the relevant context or link to the task's Prompt and select Save before asking the agent to continue. Use Continue Past Work when you need to reconstruct a longer handoff.
Choose manual or automatic execution
For a manual run, open the task, check its Agent, and select Execute. The field starts with the assigned agent, or the source agent when no agent is assigned; you can choose another. This task selection is separate from the workspace's default agent. Execution requires permission to run tasks and write access to the workspace.
For repeatable intake, ask the workflow owner to configure Task created → Agent → Update task for the workspace, activate it, and save it. Create one test task through the agent, workflow, or API, inspect the workflow result, and use Review for work that needs a person to check it. Choosing the workspace's default agent does not enable this automation. See Task workflows.