Task suites for evals on the AI Agents track. A task suite is a set of cases with fixtures and graders. Include easy cases, edge cases, and cases that must refuse. A suite of only happy paths will certify a dangerous agent.
This lesson assumes you already worked through Deterministic tests.
The idea in practice
Tag cases: happy, missing data, policy refusal, tool failure. Track pass rate per tag. Do not ship a change that drops the refusal tag.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'Task suites for evals',
}
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
Ten demos you tried by hand are not a dataset. You cannot regress them. 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
Draft eight case titles, two per tag, for a travel-booking agent.
Self-check
- Say Task suites for evals 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.