The observe-plan-act loop on the AI Agents track. The loop is observe, plan, act, then observe again. The new observation must include the tool result, not only the user's first message. Without that, the model repeats the same plan.
This lesson assumes you already worked through What an AI agent is.
The idea in practice
Store each step as a record: thought, action name, arguments, and observation. Cap the loop at a small number of steps, often 8 to 15 for a support task. When the cap is hit, stop and ask a person. Do not let the loop run until the context window fills.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'The observe-plan-act loop',
}
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
Feeding only the original question back in, and dropping the tool result, makes the agent call the same tool forever. 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
Write a five-step trace on paper for 'find the latest invoice and email the total'. Mark which step is an observation.
Self-check
- Say The observe-plan-act loop 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.