CCAR-F Practice Test 14 — 15 Questions

Question 1

An architect is designing a multi-step research agent that must gather facts, draft a report, and revise it based on a critique. The team wants to reduce the blast radius of mistakes and keep each stage's context focused, rather than letting one long conversation accumulate every intermediate result. Which TWO design choices best support this goal? (Choose two.)

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

Accepted Answer

Give each stage its own focused prompt and pass only the artifacts it needs, such as the fact list into the drafting stage, instead of replaying the entire history.. Isolating stages behind focused prompts limits what each step can see and corrupt, while an explicit validation gate stops defective artifacts from propagating. Together they confine failures to a single stage and keep context tight. These two practices deliver the containment the team wants without sacrificing the agent's access to tools or knowledge.

Answer

Remove all tools from the drafting and revision stages so the agent is forced to rely only on its parametric knowledge.

Answer

Run every stage with the highest available temperature so the agent explores many alternatives and the strongest output naturally survives.

Answer

Persist the entire conversation transcript and re-send it unchanged on every stage so no historical detail is ever lost.

Question 2

During an MCP tool invocation, a user asks Claude to look up an order by its ID. The MCP server's tool handler returns a successful result, but the response payload is 12 MB of raw JSON containing nested line items, shipping events, and audit metadata. Claude's reply to the user is vague and omits key order details. Which change to the MCP tool design best addresses this?

Accepted Answer

Return only a compact, task-relevant projection of the order data from the MCP server, with pagination or a detail-fetch tool for anything further.. The tool returned successfully, so transport, timeouts, and output token limits are not the issue. The failure is context quality: a 12 MB nested payload buries the fields the user cares about and degrades the model's answer. Shrinking the tool result to a task-relevant projection, and exposing deeper data through pagination or a dedicated detail tool, restores precision without losing capability.

Answer

Raise the MCP server's request timeout so the 12 MB response can be fully streamed before Claude processes it.

Answer

Add a system prompt instruction telling Claude to always ignore audit metadata and shipping events when answering order questions.

Answer

Increase the max_tokens setting on the Claude request so the model has more room to summarize the 12 MB payload.

Question 3

An architect is designing a Claude agent that must coordinate a research subagent and a summarization subagent to produce a market brief. The research subagent gathers sources, and the summarization subagent condenses them. The architect wants to avoid the orchestrator losing track of which subagent produced which artifact across many turns. Which TWO architectural practices best support reliable artifact tracking and handoff? (Choose two.)

Accepted Answer

Give each subagent a distinct tool name and require its results to be returned as tool_result blocks tied to a unique tool_use_id.. Reliable multi-subagent handoff depends on explicit, structured provenance rather than model inference. Unique tool names and tool_use_id linkage bind each result to its requesting subagent, while metadata embedded in artifacts keeps identity attached even after summarization or storage. Raising temperature, concatenating outputs, or trusting the system prompt to remember provenance all remove or weaken those structural guarantees, making misattribution more likely across many turns.

Answer

Concatenate all subagent outputs into a single summary string before appending them to the orchestrator's history.

Answer

Rely on the orchestrator's system prompt to remember which subagent produced each artifact without adding identifiers.

Answer

Increase the orchestrator's temperature so it explores multiple attribution hypotheses before choosing one.

Question 4

You are designing a chatbot that maintains a long conversation with users over many turns. You notice that after 50 turns, Claude starts forgetting details mentioned earlier, such as the user's name and preferences. Which architectural approach best addresses this issue?

Accepted Answer

Periodically summarize the conversation history and include the summary in the system prompt for subsequent turns.. Summarizing the conversation history and injecting it into the system prompt keeps essential details alive across many turns without exceeding token limits. This approach maintains context efficiently. Simply increasing max_tokens or temperature does not address memory, and sending full history is impractical for long chats.

Answer

Send the entire conversation history in every request without modification.

Answer

Use a higher temperature to make the model more creative in recalling details.

Answer

Increase the max_tokens parameter to allow longer responses.

Question 5

An architect is onboarding a large team to Claude Code and wants to standardize how project-level configuration is shared through version control so that every developer gets the same behavior. Which two practices should the architect adopt? (Choose two.)

Accepted Answer

Commit a `.claude/settings.json` file to the repository so shared permissions, hooks, and environment settings are consistent for everyone who clones it.. Project-scoped configuration belongs in the repository so it is versioned and shared. Committing `.claude/settings.json` standardizes permissions, hooks, and environment behavior, while committing `.claude/commands` distributes reusable slash commands. Together they give every developer identical Claude Code behavior on clone, which is the goal of the onboarding standardization effort.

Answer

Commit each developer's `~/.claude/settings.json` into the repository so personal preferences travel with the code.

Answer

Store API keys and other credentials in the committed `.claude/settings.json` so all developers authenticate identically.

Answer

Have each developer run `claude config set --global` with the team's preferred values during onboarding instead of committing configuration files.

Question 6

You are using the Anthropic Claude API in a customer-support application. Your prompt asks Claude to classify each incoming ticket into exactly one of four categories and return the result. Occasionally, Claude adds a friendly sentence before the category, which breaks your parser. Which change most directly prevents this extra text?

Accepted Answer

Add a prefill to the assistant turn, such as an opening XML tag, so the response continues from that point.. Prefilling the assistant turn is the most direct control over the start of the response. By beginning the assistant message with the exact tag or token that should open the answer, the model is forced to continue from there, eliminating any preamble. Temperature and max_tokens affect randomness and length, not format, and a friendly system persona does not enforce structure.

Answer

Increase the max_tokens value so the model has room to output the category.

Answer

Set the temperature parameter to 0.0 in the API request.

Answer

Add a system prompt that says 'You are a helpful assistant for a support team.'

Question 7

An engineering team is building a Claude-powered contract analyzer that must return a JSON object with a 'parties' array, an 'effective_date' string, and a 'governing_law' string. Legal reviewers report that for about one in twenty contracts, the model invents a governing law when the contract is silent. The team wants to stop fabricated values without losing valid extractions. Which approach best addresses this?

Accepted Answer

Instruct Claude to set 'governing_law' to null when the contract does not state one, and include a few-shot example of a silent contract returning null.. Fabricated values appear when the requested schema implies every field must be populated and the model has no sanctioned way to report absence. Explicitly allowing null for missing fields, paired with a few-shot example of a silent contract, gives the model a correct representation for gaps. Rejecting documents, raising temperature, or inferring from jurisdiction knowledge all either discard good data or amplify invention instead of eliminating it.

Answer

Increase the model's temperature so it considers more possible governing laws and picks the most probable one.

Answer

Ask Claude to search its training knowledge for the governing law typically used in the contract's jurisdiction and fill that in.

Answer

Add a rule that any contract without a governing law clause should be rejected and returned to the legal team for manual review.

Question 8

A team uses Claude to generate release notes from git commit messages. The output must be a JSON array of objects with 'title' and 'summary' fields. The model sometimes wraps the JSON in markdown fences or adds a conversational preamble. Which approach most reliably yields parseable JSON?

Accepted Answer

Define a tool with an 'input_schema' describing the array of release note objects and set 'tool_choice' to that tool.. Forcing a tool call whose 'input_schema' describes the release note array makes the model's output a structured tool_use block that must conform to the schema. This removes markdown fences and conversational text because the response is not free-form. Downstream code can consume the tool call arguments as JSON directly, giving deterministic parsing.

Answer

Set 'max_tokens' to a small value so the model cannot add extra text around the JSON.

Answer

Use a system prompt that says 'Do not use markdown' and rely on Claude to follow it.

Answer

Instruct Claude to respond only with JSON and include a prefilled assistant message starting with '{'.

Question 9

An architect is building an MCP server that exposes a tool named `fetch_weather`. The tool's input schema currently defines `location` as a string with no description. During testing, the model frequently sends postal codes for cities outside the supported region, causing upstream API errors. Which change to the tool definition most directly improves the model's ability to supply valid inputs?

Accepted Answer

Add a `description` to the `location` property that states the accepted formats and supported regions.. Tool schemas are the model's contract for argument generation. When the model supplies out-of-region values, the fix is to encode the accepted formats and supported regions in the property description so the constraint is visible during inference. Adding validation tools, raising token limits, or changing the type do not communicate the regional restriction and therefore do not prevent the invalid input.

Answer

Register a second tool called `validate_location` and rely on the model to call it before `fetch_weather`.

Answer

Increase the max_tokens setting on the Claude API request so the model has more room to reason about the location.

Answer

Change the `location` type from string to number so the model is forced to send a numeric postal code.

Question 10

A platform team maintains a repository where a committed .claude/settings.json grants broad Bash permissions, but an engineer needs a stricter personal policy that still allows the shared project settings for everything else. Which approach correctly applies the engineer's narrower permission policy?

Accepted Answer

Add the narrower rules to .claude/settings.local.json in the repository working tree. Claude Code merges settings from multiple scopes, and the local settings file exists precisely for per-developer overrides that should not be committed. Writing stricter permission rules into .claude/settings.local.json lets the engineer tighten their own environment while the committed project settings continue to apply for shared concerns, achieving a personal policy without disturbing teammates or the repository.

Answer

Create a CLAUDE.md file containing the permission rules as natural-language instructions

Answer

Edit the committed .claude/settings.json to remove the broad permissions, then commit the change

Answer

Set the CLAUDE_CODE_PERMISSIONS environment variable in the shell profile to override the project file

Question 11

A financial analyst uses Claude to extract key figures from quarterly earnings reports. The reports often contain tables with merged cells and footnotes. The analyst reports that Claude sometimes misattributes a number to the wrong quarter. You need to improve reliability without changing the model. Which technique is most likely to reduce misattribution?

Accepted Answer

Pre-process the report to convert tables into a structured format like CSV with clear headers, then ask Claude to extract from that.. Misattribution of figures to quarters often stems from ambiguous table layouts. Pre-processing the report into a structured format with clear headers eliminates that ambiguity, allowing Claude to extract values accurately. While chain-of-thought and transcription can help in some cases, they do not address the root cause as directly. Increasing max_tokens is irrelevant to parsing accuracy.

Answer

Increase the max_tokens parameter to ensure the entire report is processed in one go.

Answer

Ask Claude to think step by step before providing the extracted figures.

Answer

Provide the report as a high-resolution image and ask Claude to transcribe the tables into markdown before extraction.

Question 12

An MCP server wraps a ticketing API. The API accepts a `priority` field whose valid values are `low`, `medium`, and `high`. During testing, the model occasionally sends `urgent`, and the API returns a generic 400 error. The architect wants the model to self-correct on the next turn without changing the upstream API. Which change to the MCP tool best achieves this?

Accepted Answer

Add an `enum` of `low`, `medium`, and `high` to the `priority` field in the tool's input schema.. Encoding the allowed values as an enum in the tool schema is the cleanest fix because it prevents invalid submissions and, when a call is still rejected, gives the model an explicit set to choose from on the retry. Silently coercing values hides errors, and passing raw upstream errors through leaves the model guessing. The schema is the right place for this contract.

Answer

Map any unrecognized priority to `medium` inside the server before forwarding the request.

Answer

Return the upstream 400 error text verbatim to the model and let it infer the allowed values.

Answer

Add a free-text `notes` field where the model can explain its intended priority in prose.

Question 13

A platform team wants Claude Code to automatically run the project's formatter after every file edit made by the assistant, so that generated code always matches the repository style without manual intervention. Which configuration should the team use?

Accepted Answer

Define a `PostToolUse` hook in `.claude/settings.json` that matches the file-editing tool and invokes the formatter command.. Claude Code hooks are configured in settings files and bind shell commands to lifecycle events. A `PostToolUse` hook that matches the file-editing tool and runs the formatter guarantees the formatter executes after each edit, which is the deterministic behavior the platform team wants for repository-wide style consistency.

Answer

Set the `CLAUDE_CODE_AUTO_FORMAT` environment variable to `true` in the shell profile before launching Claude Code.

Answer

Create a custom slash command named `/format` and instruct developers to run it after each assistant edit.

Answer

Add the formatter invocation to the `CLAUDE.md` memory file as an instruction the assistant should follow after each edit.

Question 14

An architect is preparing an MCP server for a code review tool. The tool 'review_code' takes a 'language' parameter and a 'code_snippet' parameter. The architect wants to ensure the model provides valid language values and that the snippet is not excessively large. Which combination of JSON Schema constraints should be applied to the tool definition?

Accepted Answer

Use an enum for 'language' and a maxLength constraint for 'code_snippet'.. Using an enum for 'language' restricts the model to supported values, and a maxLength constraint on 'code_snippet' bounds the input size. These are standard JSON Schema keywords that directly address the requirements. They are simple, declarative, and enforced by the MCP server, making them the most appropriate choice.

Answer

Use a numeric type for 'language' and a maximum value for 'code_snippet'.

Answer

Use an array of strings for 'language' and an object type for 'code_snippet'.

Answer

Use a regex pattern for 'language' and a minLength constraint for 'code_snippet'.

Question 15

An architect is building an MCP server that exposes a tool for querying an PostgreSQL database. The tool's input schema defines a single property 'query' of type string. During testing, the model generates SQL statements that include multiple statements separated by semicolons, causing unintended data modifications. Which change to the tool design best mitigates this risk while preserving the tool's usefulness?

Accepted Answer

Parse the SQL string and reject any input containing more than one statement before execution.. The correct approach is to validate the SQL input at the MCP server and reject any input that contains more than one statement. This enforces a strict single-statement policy regardless of how the model formats its request. It keeps the tool flexible for legitimate single-statement queries while eliminating the risk of stacked statements that could modify data unexpectedly.

Answer

Add a separate boolean input property 'allowMultipleStatements' defaulting to false.

Answer

Change the tool to accept a structured object with separate properties for table, columns, and where clause.

Answer

Change the 'query' property to an enum of pre-approved SQL statement templates.