Use case · Operations
Agent teams.
Several agents work one case together. Each one has its own job and its own tools.
Hard cases get split the way a real team would split them. Every hand-off is recorded, so you can see who did what.
How it works
Step by step.
Press play, or click through it yourself. Try the switch above the drawing.
Try it · The facts
the casecoordinatorresearchercheckerwriterreply
new case
01A case comes in. It needs several kinds of work.
01 / 06
Where it's used
One idea, many industries.
- 01InsuranceOne agent gathers the claim facts. One checks them against the policy. One drafts the letter.
- 02ResearchA researcher agent collects sources. A writer agent turns them into a brief.
- 03Customer serviceA triage agent hands each case to the agent that knows that product.
- 04Due diligenceAgents split a data room by topic and report back to one coordinator.
How Promptise makes it work
All three parts, working together.
In code
A peek at the code.
- 01Each peer becomes a tool the coordinator can call.
- 02Peers are ordinary agents with their own tools.
- 03Delegation has timeouts and a depth limit.
team.pypython
from promptise import build_agentfrom promptise.cross_agent import CrossAgent coordinator = await build_agent( model="openai:gpt-5-mini", instructions="Split the case. Ask the researcher, then the writer.", cross_agents={ "researcher": CrossAgent(agent=researcher, description="Collects facts"), "writer": CrossAgent(agent=writer, description="Writes the reply"), },)