Tasks that take many steps on the AI Agents track. Multi-step work needs a goal, a scratchpad, and a budget. The scratchpad holds facts already confirmed so the model does not re-ask. The budget is steps, tokens, and money.
This lesson assumes you already worked through Structured tool calls.
The idea in practice
After each step, update a 'known facts' list and a 'still open' list. Stop when open is empty or the budget is spent. Show both lists in the log.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'Tasks that take many steps',
}
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
Restarting from the original prompt every step forgets confirmed facts and repeats work. 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
For a three-tool task, write the known-facts list after step two and what is still open.
Self-check
- Say Tasks that take many steps 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.