CCDV-F Practice Test 9 — 10 Questions

Question 1

A developer is designing a prompt that asks Claude to extract action items from meeting transcripts. The transcripts are noisy, with overlapping speakers and side conversations. Which TWO prompt-engineering practices will most improve the reliability of the extracted action items? (Choose two.)

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Instruct the model to think step by step internally before emitting the final list, then output only the list.. Few-shot examples anchor the noisy-input-to-clean-output mapping, and asking for internal reasoning before the final list lets the model resolve attribution and filter side talk. Together they address the two hard parts of this task: deciding what counts as an action item and deciding who owns it. Output limits and label removal do not improve either decision.

Answer

Remove all speaker labels to reduce token count before sending the transcript.

Answer

Set the max_tokens parameter as high as possible so the model never truncates an action item.

Answer

Ask the model to include every sentence that mentions a task, regardless of who said it.

Question 2

A developer notices that a Claude integration's monthly bill is far higher than expected. Logs show that each request includes a 15,000-token system prompt containing detailed style guidelines, plus a 200-token user question, and the model generates a 50-token answer. The system prompt is identical across all requests. Which change would MOST reduce cost without altering output quality?

Accepted Answer

Enable Prompt Caching on the static system prompt so repeated requests pay a reduced cache-read rate for those tokens.. The dominant cost driver is the 15,000-token system prompt repeated on every request. Since it is static, Prompt Caching reduces the billed rate for those tokens on subsequent calls while keeping the prompt intact. This preserves output quality exactly and targets the largest cost component, making it the most effective change.

Answer

Shorten the system prompt by removing style guidelines that the model already follows by default.

Answer

Move the user question before the system prompt so the model processes it first.

Answer

Switch to a model with a smaller context window to force the system prompt to be truncated automatically.

Question 3

An enterprise uses Claude to generate SQL queries from natural-language questions against a production database. The security team wants to prevent the model from producing destructive statements such as DROP TABLE. Which control provides the strongest guarantee?

Accepted Answer

Execute generated SQL with a database role that has only SELECT privileges on the required views.. The strongest guarantee comes from enforcing least privilege in the database, because the database engine will reject destructive statements regardless of what the model generates. Prompt instructions and keyword filters operate before execution and can be bypassed or produce errors. Logging is only detective. By executing generated SQL with a read-only role scoped to required views, the system prevents destructive operations by design.

Answer

Add a system prompt that instructs Claude to only generate SELECT statements.

Answer

Parse the generated SQL and reject any statement containing the word DROP or DELETE.

Answer

Log all generated SQL statements and alert the security team when destructive keywords appear.

Question 4

A developer is creating a multi-agent system where a coordinator agent delegates subtasks to specialized agents. Each specialized agent has its own tools and system prompt. The coordinator needs to know when a subtask is complete and what result was produced. Which design best fits the Anthropic SDK's stateless model?

Accepted Answer

Have the coordinator invoke each specialized agent as a tool, passing the subtask as input and receiving the agent's final result as the tool output.. Because each API call is independent, delegation is best modeled as a tool call: the coordinator requests a sub-agent by name, the application runs it and returns the result as tool output. This creates an explicit contract for completion and results, and it keeps each specialized agent's tools and system prompt isolated.

Answer

Have the coordinator and specialized agents share a single conversation history that all agents append to directly.

Answer

Have all specialized agents run in parallel and write their outputs to a shared memory store that the model reads automatically.

Answer

Have the coordinator emit a plain text message describing the subtask and rely on the specialized agent to detect it by polling.

Question 5

A developer is building an internal Claude-powered assistant using the Anthropic API. The assistant must access a proprietary knowledge base stored in an Amazon S3 bucket. The security team requires that the assistant never receives long-lived AWS credentials and that access is tightly scoped to only the necessary S3 prefix. Which approach best meets these requirements?

Accepted Answer

Configure the application to assume an IAM role with a scoped policy and retrieve temporary credentials via AWS STS, then use those credentials only within the tool's execution context.. The correct approach uses temporary credentials from AWS STS via an assumed IAM role. This ensures no long-lived keys exist, access is scoped to the required S3 prefix through the role's policy, and credentials are only present during tool execution. Static keys or public access fail the requirements and increase risk.

Answer

Embed the S3 bucket's public URL in the tool code and rely on the bucket being publicly readable for the specific prefix.

Answer

Store IAM user access keys in environment variables on the application server and pass them to the tool that reads S3.

Answer

Create a separate IAM user for the assistant and attach the AmazonS3FullAccess managed policy so it can read any bucket.

Question 6

A developer maintains a Claude-powered assistant that answers questions over a 60,000-token internal policy corpus. The corpus is stable and reused across every request, and the developer wants to cut cost and latency while preserving answer quality. Which two changes will most directly reduce per-request token processing for this workload? (Choose two.)

Accepted Answer

Retrieve only the policy sections relevant to each question and place those sections in the prompt instead of the full corpus.. Prompt caching avoids reprocessing the unchanged corpus prefix, and retrieval shrinks the context to only the passages a given question needs. Both act directly on the number of tokens the model must process per request, and both preserve quality because the relevant policy content remains available to the model.

Answer

Raise the temperature so Claude explores a wider range of possible answers and settles faster.

Answer

Add a system instruction telling Claude to read the corpus more quickly to save tokens.

Answer

Convert the corpus to a single long paragraph with no headings to reduce whitespace tokens.

Question 7

A developer is implementing an MCP client that connects to a remote MCP server over HTTP. The server requires authentication using an API key. The developer wants to ensure that all requests from the client include the API key in the headers. Which approach is correct for configuring the MCP client?

Accepted Answer

Set the API key in the 'Authorization' header when establishing the HTTP connection to the MCP server.. The correct approach is to set the API key in the 'Authorization' header of the HTTP connection. MCP over HTTP relies on standard HTTP authentication mechanisms, and headers are the appropriate place for credentials. Other methods either misuse query parameters, confuse client and server roles, or attempt to embed authentication in the JSON-RPC body, which is not standard.

Answer

Include the API key in the body of each MCP JSON-RPC request.

Answer

Pass the API key as an environment variable when starting the MCP server process.

Answer

Include the API key as a query parameter in the MCP server URL.

Question 8

A developer is adding a 'calculate_shipping' tool to an existing MCP server used by a Claude-based checkout assistant. The tool needs the destination country, package weight in kilograms, and an optional express flag. Which tool definition practice will MOST reliably let Claude call the tool correctly?

Accepted Answer

Define an input schema with typed properties for country, weight_kg, and express, mark the first two as required, and write a description that explains units and when to use the tool.. Tool definitions work best when the schema encodes structure and the description encodes intent. Typed properties for country, weight_kg, and express, with only the truly necessary fields required, give Claude an unambiguous contract. Documenting units and usage in the description helps the model choose the tool and format arguments correctly, which is the most reliable path to accurate invocations.

Answer

Omit the input schema entirely and rely on the tool name 'calculate_shipping' to imply the required arguments.

Answer

Set every property as required, including express, and instruct Claude to pass false when the user does not mention express shipping.

Answer

Accept a single free-form 'details' string and parse it inside the tool handler, since Claude writes natural language well.

Question 9

A developer is creating an agent that uses the Anthropic SDK to answer customer support questions. The agent has access to a tool that retrieves order status from an internal API. During testing, the developer notices that the agent sometimes provides an order status without calling the tool, resulting in incorrect information. What is the most likely cause of this behavior?

Accepted Answer

The tool's description is too vague, so the model does not understand when it should be used.. The most likely cause is a vague tool description. The model uses the description to understand the tool's purpose and when to invoke it. If the description does not clearly explain that the tool retrieves order status, the model may answer from its own knowledge, leading to incorrect information. Improving the description will guide the model to use the tool appropriately.

Answer

The agent's system prompt does not include the current date, so the model cannot determine order freshness.

Answer

The `tool_choice` parameter is set to `{"type": "auto"}`, which allows the model to ignore tools.

Answer

The model's temperature is set too high, causing it to hallucinate tool calls.

Question 10

A developer sends a Messages API request that includes a system prompt and several prior turns, and Claude replies with stop_reason set to "max_tokens" even though the conversation is short. They want Claude to finish its thought rather than truncate. Which change most directly addresses the truncation?

Accepted Answer

Increase the max_tokens value in the request so Claude has more room to complete its response.. The stop_reason value max_tokens is a direct signal that the output token limit was reached before the model finished. The remedy is to raise max_tokens within the model's context constraints; stop sequences, temperature, and prompt placement do not affect how many output tokens are permitted.

Answer

Move the instructions from the system prompt into the first user turn to free output tokens.

Answer

Lower the temperature to 0 so the response becomes deterministic and naturally shorter.

Answer

Add a stop_sequence that matches the end of the expected answer so Claude knows when to stop.