Skip to content

Prompt and Context Engineering Questions

Practice questions for Prompt and Context Engineering topic in Claude Certified Developer – Foundations. 22 questions covering this domain.

22 questions4 easy10 medium8 hard
Q1
medium

A developer wants to guarantee that Claude always returns a valid JSON object matching a specific schema, even when the prompt instructions alone resu...

Q2
easy

In a Claude API request, where should persistent developer instructions, personas, and context that should apply throughout the entire conversation be...

Q3
medium

A developer wants to clearly separate three sections in a prompt: input document, extraction rules, and response format. Which structural technique do...

Q4
hard

A developer adds cache_control to a 2,000-token system prompt for Claude Haiku 4.5. After multiple requests, the response usage always shows cache_cre...

Q5
hard

A banking chatbot with the system prompt "Answer only questions about banking" correctly refuses off-topic requests during QA testing. In production, ...

Q6
medium

A customer service chatbot has been running a 3-hour session and the context window is now 90% full. The developer wants to continue the session witho...

Q7
medium

A developer has a system prompt (5,000 tokens) containing static instructions and includes a per-request research paper (30,000 tokens) followed by th...

Q8
hard

A developer runs a two-step code generation pipeline: step 1 generates a design plan, step 2 generates implementation code. The plan generation (step ...

Q9
medium

A developer's Claude-powered pipeline produces inconsistent JSON output about 3% of the time despite a system prompt instruction saying "always return...

Q10
hard

A developer uses explicit cache breakpoints. After 10 conversation turns (10 total blocks), they place cache_control on the last message. The conversa...

Q11
easy

Which technique involves beginning Claude's response with specific text to guide its output format, often called "assistant turn prefilling"?

Q12
hard

A developer builds a contract analysis agent with a 10,000-token system prompt containing detailed legal reasoning guidelines. The agent processes 200...

Q13
hard

A developer implements a multi-pass review workflow where Claude first generates a code implementation, then reviews its own implementation in the sam...

Q14
medium

A developer builds a document extraction pipeline where Claude must return JSON with specific fields. The system prompt includes detailed JSON schema ...

Q15
medium

A legal research assistant uses Claude to analyze contracts. The system prompt is 2,000 tokens. Each contract is 30,000 tokens. The application proces...

Q16
easy

Which prompting technique involves providing Claude with several example input-output pairs before the actual task to guide its response format and st...

Q17
medium

A developer wants to ensure that sections of user-provided content are clearly distinguishable from system instructions in the prompt to reduce prompt...

Q18
medium

A legal document analysis system includes a 40,000-token system prompt with detailed legal reasoning frameworks. The development team notices inconsis...

Q19
easy

Which prompt engineering technique involves providing Claude with explicit examples of desired outputs before presenting the actual task?

Q20
medium

A developer builds a pipeline where Claude first extracts entities from documents (step 1), then classifies relationships between entities (step 2). S...

Sign in to see all 22 questions

Create a free account to browse the questions included with the free plan.