Course Test

Please answer the questions below to complete the test.

Question 1

Which of the following best describes what a coding agent can do that a standard chat LLM cannot?

Select one answer

Generate longer responses with more detail.

Call tools, execute commands, and iterate across multiple steps to complete a task.

Access the internet in real time.

Use a different underlying model architecture.

Question 2

In the agentic loop, what is the purpose of the 'observe' step?

Select one answer

To pause the loop and wait for human approval.

To record the action in an audit log.

To append the tool result to the agent's context so it can plan the next step.

To determine whether the step limit has been reached.

Question 3

Which of the following are valid reasons to terminate an agentic loop before the task is complete?

Select all that apply

The maximum number of allowed steps has been reached.

The agent has produced an intermediate result but has not finished.

The agent cannot proceed without information or permission it does not have.

The agent has called at least one tool.

Question 4

What is the primary reason to use `kubectl apply --dry-run=server` before an agent applies a manifest to a cluster?

Select one answer

To speed up the apply operation on the next run.

To validate the manifest against the API server without persisting any change.

To create a backup of the current cluster state.

To notify the human operator that a change is about to be applied.

Question 5

Which of the following best describes 'blast-radius thinking' in the context of agent permissions?

Select one answer

Estimating how fast the agent can complete its task.

Calculating the number of tool calls in the agentic loop.

Assessing the worst-case impact of a tool access and constraining permissions accordingly.

Measuring the agent's confidence score before it takes an action.

Question 6

Which Meshery command would you run to check that Meshery's adapters and cluster connectivity are healthy before delegating infrastructure tasks to an agent?

Type your answer below