When several agents help on the AI Agents track. Several agents help when roles are truly different, such as researcher and checker. They hurt when they all share one vague goal and talk in circles. Start with one agent. Split only after a trace shows one role drowning.
This lesson assumes you already worked through Research workflows.
The idea in practice
Give each agent a contract: inputs it accepts, outputs it must return, tools it may call. A coordinator passes structured outputs, not a group chat.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'When several agents help',
}
checks = [
'input available at decision time',
'score matches the real decision',
'failure case written down',
]
print(goal['lesson'])
for item in checks:
print('-', item)
Run the sketch locally if you have Python. The printout is a reminder of the checks, not a trained model. Replace the strings with the real inputs from your own example before you treat it as a design.
What usually goes wrong
A group chat of three general agents triples cost and hides which one was wrong. When this happens, stop adding parameters or tools. Fix the check, the data, or the permission, then run the same example again.
What to write down
- The input you are allowed to use at decision time.
- The output and the score or pass rule.
- One failure you will test on purpose.
- What you will not claim the system can do.
Practice
Name a task you would keep as one agent, and one you would split into writer and checker. Say why.
Self-check
- Say When several agents help in one sentence that mentions an input and an output.
- Name the failure mode in this lesson and the check that would catch it.
Done when: you can explain this lesson without the page open, and you have a written failure case.