Boundaries around actions on the AI Agents track. Guardrails are checks outside the model: allow-lists, max amounts, rate limits, and banned tools. The model can propose a refund of any size. The guardrail refuses amounts above policy.
This lesson assumes you already worked through Human approval gates.
The idea in practice
Run guardrails after schema validation and before the tool. Return a clear refusal observation. Do not rely on the system prompt as the only limit.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'Boundaries around actions',
}
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 prompt that says 'never refund over $50' is not a control. The model can ignore it. 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 three guardrails for a support agent: amount, recipient, and how many refunds per hour.
Self-check
- Say Boundaries around actions 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.