CCDV-F Practice Test 17 — 15 Questions

Question 1

Why should you avoid including sensitive personal data (PII) in the tool results returned to an agent?

Accepted Answer

It increases the risk of data leakage via the model's history.. PII returned to an agent becomes part of the conversation history, which is often logged or cached for model context. This increases the risk of inadvertent data exposure. By scrubbing PII before passing tool results to the model, developers maintain security compliance and ensure that sensitive data is not unnecessarily processed or retained by AI infrastructure, aligning with enterprise data protection standards.

Answer

It significantly slows down the model inference time.

Answer

It reduces the model's reasoning capabilities.

Answer

The model cannot understand PII formats.

Question 2

Claude requests a tool call to 'query_database', but the database is currently down. What is the best practice for handling this error so Claude can inform the user effectively?

Accepted Answer

Send a 'tool_result' with is_error: true and a descriptive error message.. Graceful error handling is vital for a good user experience. Instead of failing the entire API request, the developer should return the error message within the 'tool_result' block. By setting 'is_error' to true, the developer signals to Claude that the tool failed, allowing the model to explain the situation to the user.

Answer

Return a 500 Internal Server Error to the Anthropic API request.

Answer

Omit the 'tool_result' and send a new user message saying 'The tool failed'.

Answer

Modify the assistant's previous tool_use block to remove the request.

Question 3

When designing prompts for complex reasoning tasks, which TWO practices are recommended by Anthropic to improve the reliability of the output?

Accepted Answer

Instructing the model to output its reasoning step-by-step inside <thinking> tags.. Complex reasoning requires structural guidance and transparency in the model's internal logic. By encouraging the model to think before answering and providing clear delimiters for input data, developers can significantly reduce errors. These techniques ensure that the model processes information linearly and allocates sufficient computational focus to the logic before committing to a final answer.

Answer

Setting the top_p parameter to 1.0 to ensure the widest possible range of reasoning paths.

Answer

Placing the most important instructions in the middle of a very long prompt to avoid bias.

Answer

Combining multiple unrelated tasks into a single prompt to maximize token efficiency.

Question 4

A developer needs to monitor the costs of different departments using a single Anthropic API key. Which API feature should be utilized to categorize and track usage without creating multiple accounts or keys?

Accepted Answer

The 'metadata' object with a 'user_id' or custom field.. Tracking usage across different business units is a common requirement for cost management. Anthropic provides the 'metadata' field in the Messages API, allowing developers to attach custom identifiers like 'department_id'. These tags appear in usage logs and billing exports, enabling precise cost allocation and budget monitoring without the administrative overhead of managing multiple API keys.

Answer

The 'system' prompt field.

Answer

The 'stop_sequences' parameter.

Answer

The 'top_p' sampling parameter.

Question 5

Which of the following is an example of an effective natural language instruction for Claude Code?

Accepted Answer

Refactor the authentication module to use JWT instead of sessions, keeping existing error handling.. Effective prompts are clear, concise, and provide actionable goals with necessary context. By specifying the file, the intended change, and the desired outcome, a developer helps the agent narrow its focus, leading to more accurate results. This practice reduces ambiguity, minimizes the need for follow-up corrections, and maximizes the efficiency of the AI agent, which is a key skill for any developer working with modern LLM-powered CLI tools.

Answer

Fix the code.

Answer

Make it better.

Answer

Delete everything in the src directory.

Question 6

Refer to the exhibit. Claude has generated two 'tool_use' blocks in a single response. Which approach is valid for handling these calls in the client application?

Accepted Answer

The client can execute them in parallel and must return two tool_result blocks.. Claude supports parallel tool use, allowing it to request multiple operations at once to improve efficiency. The client can execute these calls in parallel or sequentially, but it must return a separate tool_result block for each tool_use block, ensuring the 'tool_use_id' in the result matches the 'id' provided by the model.

Answer

The client must execute them sequentially to avoid race conditions in the API.

Answer

The client should combine both results into a single tool_result block.

Answer

The client must discard the second call and only process the first one.

Question 7

Refer to the exhibit. A developer provides this tool definition to Claude. If the user asks for the weather in 'London' without specifying a unit, how will Claude's tool_use block be structured based on this JSON schema?

Accepted Answer

It will include only the 'location' parameter in the input object.. Claude follows the provided JSON schema strictly when generating tool calls. Since the 'unit' property is not included in the 'required' array, Claude is not obligated to provide it if the user does not specify it. This behavior demonstrates Claude's ability to handle optional parameters while adhering to the structure defined in the tool's input_schema field.

Answer

It will fail with a validation error because 'unit' is missing.

Answer

It will prompt the user to choose between 'celsius' and 'fahrenheit' first.

Answer

It will default the 'unit' to 'celsius' automatically.

Question 8

Your organization is integrating Claude for sensitive internal human resources queries. What is the most effective way to ensure the model does not reveal employee salary information?

Accepted Answer

Implement a data-masking middleware that filters sensitive fields from the retrieved context.. The most secure way to prevent sensitive data disclosure is to prevent the model from ever 'seeing' that data during the generation phase. By implementing a logic layer that filters the information based on the user's identity before it reaches the model, you ensure that the model is only provided with authorized context. This prevents the model from acting as an unauthorized channel for sensitive PII or payroll data.

Answer

Instruct the model in the system prompt to never discuss salaries.

Answer

Require the model to perform a sanity check on its own output for PII.

Answer

Encrypt all employee salaries using a salt and hash.

Question 9

A developer is using the 'forced tool use' feature. What will be the value of the 'stop_reason' field in the API response when Claude generates the tool call?

Accepted Answer

tool_use. The 'stop_reason' field indicates why the model stopped generating text. When the model is forced to use a tool, or when it naturally decides to use one, it stops its current generation turn to allow the client to execute the tool. The value 'tool_use' is the specific indicator for this state.

Answer

end_turn

Answer

max_tokens

Answer

stop_sequence

Question 10

How does Claude Code maintain context over a long-running conversation in the terminal?

Accepted Answer

It keeps state in memory and local configuration.. Claude Code utilizes a stateful management system that tracks the conversation history, recent file changes, and the current project state. This allows the model to reference previous suggestions, understand the current file structure, and build upon earlier tasks, which is critical for consistent, multi-turn interactions. By managing this context effectively, the tool provides a cohesive experience that mimics a human pair programmer, ensuring that the agent remains helpful throughout complex, multi-stage development projects.

Answer

It stores all conversation history in the git index.

Answer

It clears all history every time a command is run.

Answer

It relies entirely on the terminal scrollback buffer.

Question 11

An application uses Claude 3.5 Sonnet and implements Prompt Caching for a 10,000-token system prompt. The application processes 1,000 requests per hour. If the cache is refreshed every request and never expires, how does the billing for the input tokens change after the very first request?

Accepted Answer

The first request incurs a 'cache write' fee, and the next 999 incur 'cache hit' fees.. Prompt Caching introduces a two-tier pricing model for input tokens. The first request 'writes' the tokens to the cache, which is billed at a 'cache write' rate (usually 25% more than standard input). All subsequent requests that hit this cache are billed at the 'cache hit' rate, which is about 10% of the standard cost. This shift dramatically lowers the long-term cost of the application.

Answer

The first request is free, and subsequent requests are full price.

Answer

Every request is billed at a flat 50% discount.

Answer

Billing remains the same as standard input for all requests.

Question 12

A developer wants to use Claude for a creative writing assistant that generates unpredictable and varied story ideas. Which parameter should they primarily adjust to achieve this?

Accepted Answer

Temperature, by increasing it toward 1.0.. Temperature is the primary parameter for controlling the randomness of the model's output. Higher values (up to 1.0) make the probability distribution of the next token flatter, allowing the model to choose less likely words, which results in more creative, varied, and sometimes surprising content, which is ideal for brainstorming and storytelling.

Answer

Max tokens, to ensure the stories are long enough to be creative.

Answer

Stop sequences, to prevent the model from ending the story too early.

Answer

System prompt, by adding more technical constraints to the writing style.

Question 13

You are debugging an agent that often fails to format its final response as JSON, despite being instructed in the system prompt. Which technique is most effective for ensuring consistent output structure?

Accepted Answer

Utilizing a tool to force the output into a structured schema.. Forcing the output structure through constrained generation is the most robust way to ensure valid JSON. While system prompts set expectations, they are soft constraints. Using tool-use or structured output features allows the agent to adhere to a schema, which guarantees the output is machine-readable and prevents parsing errors in downstream applications.

Answer

Adding a 'Please provide JSON' phrase at the end of the prompt.

Answer

Lowering the system prompt length to reduce ambiguity.

Answer

Retrying the call automatically until valid JSON is returned.

Question 14

When sharing state between multiple agent turns, why is it recommended to use a managed session object instead of appending everything to a simple list of messages?

Accepted Answer

It enables intelligent context pruning and summarization.. Managed state objects allow for intelligent context truncation and summarization. In complex agent workflows, message histories can quickly exceed the model's context window. A robust session manager handles pruning older turns, maintaining critical environment variables, and ensuring the model receives only the most relevant information for the current task, which is critical for long-running autonomous processes.

Answer

It guarantees that all messages are stored in a database.

Answer

It automatically encrypts all user inputs for privacy.

Answer

It makes the model faster by skipping validation.

Question 15

What role do 'project-aware' features play in Claude Code during a development session?

Accepted Answer

They enable the agent to understand existing project conventions.. Project-awareness allows Claude Code to understand the structure, language conventions, and dependency relationships within the codebase. This context is crucial for generating code that is idiomatic and structurally sound. Instead of treating files as isolated units, the agent can reason about how changes in one file impact another, leading to more accurate refactoring and fewer integration issues across the entire project repository.

Answer

They allow the agent to automatically commit changes to Git.

Answer

They force the agent to only use the most popular libraries.

Answer

They prevent the agent from accessing external online resources.