Reflection and self-critique pattern in Agentic AI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to initialize the agent's reflection step.

Agentic AI

agent.reflection = [1]()

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

AReflectionStep

BObservationStep

CActionStep

DCritiqueStep

Attempts:

3 left

💡 Hint

Common Mistakes

Using CritiqueStep instead of ReflectionStep.

Confusing ActionStep with reflection.

Forgetting to initialize the reflection attribute.

The ReflectionStep class is used to enable the agent to reflect on its actions and outcomes.

2fill in blank

medium

Complete the code to add a self-critique after the agent's action.

Agentic AI

agent.add_step([1]())

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

AReflectionStep

BActionStep

CSelfCritiqueStep

DObservationStep

Attempts:

3 left

💡 Hint

Common Mistakes

Using ReflectionStep instead of SelfCritiqueStep.

Adding ActionStep instead of critique.

Skipping the critique step.

SelfCritiqueStep is the step where the agent evaluates its own previous action to improve.

3fill in blank

hard

Fix the error in the code to correctly perform reflection and critique.

Agentic AI

result = agent.perform_step([1])

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

ACritiqueStep

BSelfCritiqueStep()

CReflectionStep()

DActionStep()

Attempts:

3 left

💡 Hint

Common Mistakes

Using CritiqueStep without parentheses causes an error.

Passing class name instead of instance.

Using wrong step type.

SelfCritiqueStep() must be instantiated with parentheses to create an object before performing the step.

4fill in blank

hard

Fill both blanks to create a dictionary that stores reflection and critique results.

Agentic AI

results = {
  'reflection': agent.[1](),
  'critique': agent.[2]()
}

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

Areflect

Bself_critique

Caction

Dobserve

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'action' or 'observe' methods instead of reflection or critique.

Mixing up the method names.

Forgetting parentheses to call methods.

The agent.reflect() method returns reflection results, and agent.self_critique() returns critique results.

5fill in blank

hard

Fill all three blanks to define a function that runs reflection, critique, and returns combined feedback.

Agentic AI

def run_feedback(agent):
    reflection = agent.[1]()
    critique = agent.[2]()
    return {'reflection': reflection, 'critique': critique, 'summary': [3]

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

Areflect

Bself_critique

Creflection + ' and ' + critique

Dreflection + critique

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong method names.

Concatenating without spaces or connectors.

Forgetting to call methods with parentheses.

The function calls reflect() and self_critique(), then combines their strings with ' and ' for a readable summary.