CCAR-F Practice Test 8 — 10 Questions

Question 1

A support agent running on the Claude Messages API must look up a customer's order status before answering. The developer writes a single request that includes the get_order_status tool definition and the user's question, then expects the model to return the order status directly in its first response. What actually happens, and what must the developer implement?

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

Accepted Answer

Claude returns a tool_use block requesting get_order_status; the application must execute the tool and send back a tool_result, then call the API again to get the final answer.. Tool use in the Claude Messages API is a caller-driven loop: the model emits a tool_use request, the application executes the function and returns a tool_result, and a follow-up request lets Claude produce the final answer. Because the API is stateless and never runs customer functions itself, the developer must implement this round trip.

Answer

Claude returns the order status immediately because tool definitions are executed server-side by Anthropic before the response is generated.

Answer

Claude executes the tool and returns both the tool output and its final answer in one response, so no second request is needed.

Answer

The request fails with a validation error because a tool definition cannot be supplied alongside a normal user text message.

Question 2

A developer builds a Claude-powered assistant that answers questions over a 300-page policy manual. When the whole manual is placed in a single prompt, answers to questions about clauses in the middle of the document are frequently wrong or vague, while answers about the opening and closing sections are accurate. The manual must stay available in full, and the team wants to keep latency reasonable. Which technique most effectively improves accuracy for the mid-document clauses?

Accepted Answer

Split the manual into sections, embed each section, retrieve the most relevant sections for the question, and place only those retrieved sections in the prompt before the question.. Accuracy drops for material buried in the middle of a very long context because attention is spread thin across the document. Retrieving the sections most relevant to each question and placing only those in the prompt shrinks the context to on-topic passages, which restores accuracy for mid-document clauses while the full manual stays indexed for availability. Reordering or enlarging the same long context does not fix the attention problem.

Answer

Increase max_tokens substantially and instruct the model to think step by step before answering each question.

Answer

Ask the user to restate their question using exact phrases from the manual so the model can match keywords directly.

Answer

Keep the full manual in the prompt but move it so the manual text appears after the question rather than before it.

Question 3

A team is extracting medication names and dosages from clinical notes. The notes contain many similar-looking terms, and the team needs high precision to avoid false positives. They plan to use few-shot examples. Which design of the few-shot examples best supports high precision in this scenario?

Accepted Answer

Include a mix of positive examples and carefully chosen hard negatives that resemble medications but should not be extracted.. High precision in a field with lookalike terms depends on teaching the model where the boundary lies. A mix of positive examples and hard negatives does that by showing both what to extract and what to reject. Positive-only, single-example, or random sampling designs do not surface the confusing cases, so they are less likely to reduce false positives in clinical extraction.

Answer

Provide many examples of correct extractions only, covering a wide range of medication names.

Answer

Randomly sample examples from the dataset without regard to difficulty or similarity.

Answer

Use a single detailed example and rely on the model to generalize to all other notes.

Question 4

A financial analyst uses Claude via the Messages API to extract line items from scanned invoices. The response must be valid JSON conforming to a fixed schema with fields invoice_number, date, and total. During testing, the model occasionally adds a trailing sentence after the JSON object or wraps the JSON in a markdown code fence. Which approach best guarantees the response parses cleanly while preserving extraction accuracy?

Accepted Answer

Instruct the model to begin the response with a tool_use block whose input_schema defines the invoice fields, then read the structured tool input from the response.. Constraining the response through a tool definition with an input_schema channels generation into the model's structured tool-calling pathway, so the extracted invoice fields arrive as an already-parsed object instead of free-form text. This eliminates markdown fences and trailing commentary while keeping the model focused on the document content. Prompt phrasing and sampling parameters reduce, but do not remove, format drift.

Answer

Set the max_tokens parameter to the exact length of the expected JSON string so the model cannot emit extra characters.

Answer

Request the JSON inside <output> tags and run a regular expression that captures everything between the opening and closing tags.

Answer

Append the phrase 'Return only JSON' to the end of the system prompt and set temperature to 0.0 before each request.

Question 5

An architect is standardizing Claude Code usage across a large engineering organization. They want to ensure that every developer's sessions automatically inherit a curated set of project conventions and that destructive shell commands are blocked unless explicitly approved. Which TWO configurations should the architect implement? (Choose two.)

Accepted Answer

Commit a CLAUDE.md file at the repository root containing the shared coding conventions and architectural notes. A committed CLAUDE.md at the repository root distributes shared conventions automatically to every session, and deny rules in the project's .claude/settings.json enforce hard blocks on destructive commands organization-wide. Manual per-developer setup and README-based expectations lack reliability and auditability, while blanket Bash auto-approval undermines the safety objective.

Answer

Instruct each developer to run /config and manually set the desired conventions on first launch

Answer

Enable auto-approval for all Bash commands to reduce interruptions during long sessions

Answer

Place convention notes in a README.md and expect Claude to infer them from repository files

Question 6

While working in a repository, an architect wants Claude Code to explain the purpose of a specific function in a large source file without altering any code or generating new files. Which mode should be used for this task?

Accepted Answer

Plan mode, because it prevents edits and focuses on producing an explanation or approach. Claude Code separates exploration from execution. Plan mode is the read-only posture intended for understanding code, answering questions, and outlining approaches; it will not modify files. When the goal is purely explanatory, choosing plan mode enforces the no-change requirement structurally instead of relying on the model's discretion.

Answer

Auto-accept edits mode, because approvals are bypassed and analysis proceeds faster

Answer

A new session started with the --dangerously-skip-permissions flag

Answer

Default edit mode, because Claude Code will simply describe the function without touching files

Question 7

A developer working in a monorepo wants Claude Code to follow the team's TypeScript conventions only when editing files under the `packages/api` directory, while letting other directories use different conventions. What is the most appropriate way to configure this in Claude Code?

Accepted Answer

Create a `CLAUDE.md` file inside `packages/api` containing the TypeScript conventions so Claude Code applies those instructions when working in that directory.. Claude Code discovers `CLAUDE.md` memory files by walking from the current directory up the tree, so a file placed inside `packages/api` is loaded exactly when the session operates there. This gives the API package its own TypeScript conventions while leaving sibling packages and the repository root untouched, matching the requirement for directory-scoped guidance.

Answer

Add the TypeScript conventions to the user-level `~/.claude/CLAUDE.md` file so they apply globally to every project the developer opens.

Answer

Set the `CLAUDE_CODE_MODEL` environment variable to a model fine-tuned on the API package's conventions before starting each session.

Answer

Add a `settings.json` entry with a `scope` field set to `packages/api` under the project's `.claude` directory to bind conventions to that path.

Question 8

An architect is building an MCP server that exposes a `read_file` tool. The tool's input schema currently defines `path` as a plain string. Security review requires that the tool never read outside a configured workspace root, and that the model receive a clear, machine-readable error when it attempts to do so. Which change to the tool design best satisfies this requirement?

Accepted Answer

Have the server resolve the requested path against the workspace root, reject traversal outside it, and return a structured error object.. The secure pattern is server-side enforcement plus a structured error. The MCP server should resolve the incoming path, confirm it stays within the workspace root, and return a typed error payload when it does not. That keeps the security boundary out of the model's control while still giving the model enough information to recover gracefully on the next turn.

Answer

Add a `root` parameter to the schema so the model can specify the allowed workspace root on each call.

Answer

Keep `path` as a string but document in the tool description that paths must stay inside the workspace root.

Answer

Constrain `path` in the JSON Schema with an `enum` of known filenames and reject any other value.

Question 9

You are building a customer-support agent with the Anthropic Messages API. The agent must call a `get_order_status` tool, and the result is required before it can decide whether to issue a refund. The tool sometimes takes several seconds to respond. Which approach best ensures the agent does not prematurely finalize its answer before the tool result is available?

Accepted Answer

After receiving a `tool_use` block, execute the tool and send a `user` message containing a `tool_result` block with the matching `tool_use_id` before requesting the next completion.. Tool use in the Anthropic Messages API is a multi-turn exchange: the model emits a `tool_use` block, the application executes the function, and the application returns a `tool_result` block tied to the same `tool_use_id`. The model cannot finalize a refund decision until that result is in context, so completing the round trip before the next completion is what prevents premature answers.

Answer

Use `stop_sequences` to halt generation whenever the model emits a `tool_use` block.

Answer

Set `tool_choice` to `{"type": "any"}` so the model is forced to call a tool on every turn.

Answer

Increase `max_tokens` so the model has room to wait for the tool response within the same completion.

Question 10

A developer is building a classification pipeline that routes support tickets into one of eight categories. They want Claude to return a category label and a confidence score between 0 and 1. The output must be parseable by a JSON parser without post-processing. They plan to use the Anthropic Messages API. Which approach best ensures the output is directly parseable JSON with the required fields?

Accepted Answer

Describe the JSON schema in the system prompt, prefill the assistant turn with `{`, and validate the response with a JSON parser.. Describing the schema in the system prompt gives Claude the keys and types to emit, and prefilling the assistant turn with `{` forces the response to begin as a JSON object with no preamble or markdown fence. Validating with a JSON parser catches rare deviations. Together these produce directly parseable output, unlike free-text formats, stop-sequence tricks, or high-temperature sampling.

Answer

Set `temperature` to 1 and rely on the model's creativity to produce varied but valid JSON.

Answer

Ask Claude to return the category and confidence in a single sentence like 'Category: X, Confidence: 0.9'.

Answer

Use the `stop_sequences` parameter set to '}' so Claude stops as soon as it finishes the JSON object.