Regression testing for agent changes in Agentic AI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to run a regression test on the agent's output.

Agentic AI

assert agent_output == [1]

Drag options to blanks, or click blank then click option'

Anew_output

Bagent_output

Ctest_result

Dexpected_output

Attempts:

3 left

💡 Hint

Common Mistakes

Comparing the output to itself instead of the expected result.

Using an undefined variable for comparison.

We compare the agent's output to the expected output to verify correctness.

2fill in blank

medium

Complete the code to log the regression test result.

Agentic AI

logger.[1]('Regression test passed')

Drag options to blanks, or click blank then click option'

Ainfo

Bdebug

Cwarn

Derror

Attempts:

3 left

💡 Hint

Common Mistakes

Using error or warn logging for passing tests.

Using debug which might be too low level.

Logging info level is appropriate for successful test messages.

3fill in blank

hard

Fix the error in the regression test function call.

Agentic AI

def test_agent():
    result = agent.run(input_data)
    assert result [1] expected_output

Drag options to blanks, or click blank then click option'

A!=

B>=

C==

D<=

Attempts:

3 left

💡 Hint

Common Mistakes

Using '!=' which checks for inequality.

Using comparison operators like '>=' or '<=' which are not suitable.

The test asserts that the result equals the expected output.

4fill in blank

hard

Fill both blanks to create a dictionary of test results with agent names and pass status.

Agentic AI

test_results = {agent_name: [1] for agent_name in agents if agents[agent_name].active [2] True}

Drag options to blanks, or click blank then click option'

Aagents[agent_name].run(input_data) == expected_output

Bagent.run(input_data)

C==

D!=

Attempts:

3 left

💡 Hint

Common Mistakes

Using '!=' instead of '==' in the condition.

Using the agent.run(input_data) without comparison in the dictionary.

We check if each agent's run output equals the expected output and filter agents that are active (== True).

5fill in blank

hard

Fill all three blanks to filter agents, run tests, and store pass status in a dictionary.

Agentic AI

results = {agent: (output := agent.run(data)) [1] expected for agent, expected in [2] if agent.active == [3]

Drag options to blanks, or click blank then click option'

A==

Bitems()

CTrue

D!=

Attempts:

3 left

💡 Hint

Common Mistakes

Using '!=' instead of '==' for comparison.

Using keys() instead of items() for iteration.

Checking if agent.active is False instead of True.

We compare output to expected with '==', iterate over items(), and check if agent is active (True).