Anthropic CCA-F (Claude Certified Architect - Foundations (CCAR-F)) Practice Test 16

Question 150

A CI pipeline runs Claude Code non-interactively for pull-request review. A downstream CI job must parse the review result reliably. Which option best meets the operational need?

  1. Run in a restricted sandbox with minimal tokens and no secret-bearing privileged steps
  2. Set job timeout, token or iteration budgets, and explicit failure behavior
  3. Use an independent check or human approval before merging
  4. Request a defined JSON output format and validate its schema in the pipeline
  5. Run Claude Code in noninteractive print mode with a bounded task prompt

Explanation

Correct Answer

D:

Structured output supports deterministic gating and avoids brittle parsing of free-form review prose. The relevant requirement is to emit structured machine-readable output, and this measure operates at that decision point.

Incorrect Answers

A:

Pull-request content is untrusted input; prompt instructions cannot substitute for runner isolation or secret scoping; this is useful for a pull request from an untrusted fork contains instructions to exfiltrate secrets, but it leaves the stated requirement to emit structured machine-readable output unmet.

B:

CI should fail visibly when budgets are exhausted rather than hang or silently return partial success; it protects against the case where a CI task might continue indefinitely under repeated tool calls, rather than the failure described here: a downstream CI job must parse the review result reliably.

C:

A model reviewing its own output is correlated with its mistakes; a distinct gate reduces unchecked self-approval; that applies when the team wants Claude's review to recommend fixes but not approve its own changes, whereas this case requires a response to a downstream CI job must parse the review result reliably.

E:

This measure fits a different trigger: the build runner has no interactive terminal for answering Claude prompts. In that setting, A headless execution path avoids UI expectations and lets CI capture the completed response as an artifact; it does not resolve the present need to emit structured machine-readable output.

Question 151

A coding assistant reviews generated patch descriptions for consistency. A reviewer cannot tell why two visually similar changes receive different labels. What is the most appropriate design choice?

  1. Add a small labeled example covering coordinated changes across multiple modules
  2. Use clearly illustrative placeholders and require current identifiers from the input
  3. Provide few-shot examples that distinguish behavior-preserving and behavior-changing patches
  4. Annotate samples with the behavior or contract difference that changes the classification

Explanation

Correct Answer

D:

The important signal is semantic change, not file size or formatting; rationale-bearing examples make that visible.

Incorrect Answers

A:

A representative complex case teaches how to summarize a unit of work that spans files without treating each edit as unrelated; that applies when the team's examples omit multi-file changes, whereas this case requires a response to a reviewer cannot tell why two visually similar changes receive different labels.

B:

Few-shot examples guide format but must not become a source of fabricated ticket IDs or commit hashes; this is useful for the assistant copies placeholder identifiers from example output, but it leaves the stated requirement to include the distinguishing condition in examples unmet.

C:

This measure fits a different trigger: some changes are wrongly described as refactors rather than bug fixes. In that setting, Contrastive labels teach the category distinction that ambiguous prose alone may not reliably convey; it does not resolve the present need to include the distinguishing condition in examples.

Question 152

Claude Code is exploring a large monorepo for a maintenance task. An engineer needs every reference to a deprecated function in source text. Which response is most reliable?

  1. Use a targeted content search with a bounded file scope
  2. Run the build or test using a controlled shell or task tool
  3. Use a file-read operation on the known path and relevant range
  4. Use a file glob or filename discovery operation

Explanation

Correct Answer

A:

Text search finds matching symbols inside files; filename discovery alone cannot identify call sites. Here it serves the explicit need to search file contents by symbol without adding a separate unverified inference.

Incorrect Answers

B:

Only an actual process invocation establishes compiler and test behavior; inspection tools cannot substitute for execution; that applies when a scripted maintenance operation must compile the project and observe test output, whereas this case requires a response to an engineer needs every reference to a deprecated function in source text.

C:

Direct reading preserves surrounding context when a file is already identified and avoids broad repository scans; this is useful for the engineer knows the exact configuration file and needs to inspect one section, but it leaves the stated requirement to search file contents by symbol unmet.

D:

This measure fits a different trigger: an engineer knows a filename pattern but not its directory. In that setting, Glob-style search identifies paths by name pattern without reading or executing each file in the repository; it does not resolve the present need to search file contents by symbol.

Question 153

A research publishing workflow uses confidence-aware human oversight. A low-risk spelling correction is supported by deterministic lint feedback. What should the team do to resolve this particular gap?

  1. Use observed error rates or rubric-based review rather than the unsupported percentage
  2. Require a verifiable source or explicit review of the unsupported claim
  3. Flag both claims and their evidence for accountable adjudication
  4. Apply the checked fix without adding a human approval queue

Explanation

Correct Answer

D:

Review resources can focus on uncertain decisions while deterministic cosmetic changes proceed under normal controls.

Incorrect Answers

A:

Language-model confidence expressions are not calibrated statistical guarantees unless validated against representative outcomes; this is useful for a model reports ninety-nine percent confidence without a calibrated evaluation process, but it leaves the stated requirement to automate the routine correction unmet.

B:

Publication quality depends on accessible supporting material, not merely on a plausible citation string; that applies when an automated report cites a source that cannot be retrieved by the reviewer, whereas this case requires a response to a low-risk spelling correction is supported by deterministic lint feedback.

C:

This measure fits a different trigger: the source evidence is contradictory and would change an important recommendation. In that setting, Conflicting high-impact evidence warrants review even if a model prefers one narrative with high confidence; it does not resolve the present need to automate the routine correction.

Question 154

An invoice-extraction workflow must recover from malformed output. An extracted total passes type validation but does not equal subtotal plus taxes. Which control is most directly relevant?

  1. Report the mismatched field and constraint, then retry within a bounded policy
  2. Run business-rule arithmetic validation and request targeted correction
  3. Mark the conflict and request human clarification with both source locations
  4. Require evidence or field-level source matching for supplier identity

Explanation

Correct Answer

B:

A valid numeric type does not prove accounting consistency; arithmetic rules catch errors outside the schema's expressiveness.

Incorrect Answers

A:

This measure fits a different trigger: a tool payload is not valid against the required JSON Schema. In that setting, Specific schema errors give the model a repair target; blindly regenerating without feedback repeats the same failure mode; it does not resolve the present need to check semantic consistency after parsing.

C:

A retry cannot resolve contradictory evidence by itself; the ambiguity belongs in review rather than arbitrary selection; this is useful for the invoice number appears in two places with different values, but it leaves the stated requirement to check semantic consistency after parsing unmet.

D:

Syntax repair does not imply factual correctness; source-backed validation checks whether each critical value is actually present; it protects against the case where a corrected output has valid JSON but an invented supplier name, rather than the failure described here: an extracted total passes type validation but does not equal subtotal plus taxes.

Question 155

A Claude Code assistant supports an engineering team during feature development. A developer wants generated code to conform to existing examples. How should the workflow be adjusted?

  1. Reference nearby implementations and conventions while specifying differences
  2. Use a structured requirement interview covering behavior and constraints
  3. Inspect the failed assertion and revise only the mismatched logic
  4. Review exported signatures and compatibility constraints as separate checks
  5. Create cases for missing, empty, and malformed inputs before coding

Explanation

Correct Answer

A:

Real repository examples anchor the solution in established architecture without forcing the model to invent project-specific patterns.

Incorrect Answers

B:

This measure fits a different trigger: the request has several plausible interpretations of the user experience. In that setting, Clarifying choices reduces wasted implementation iterations and ensures stakeholders own product decisions; it does not resolve the present need to show representative local patterns rather than generic style advice.

C:

Failure traces narrow the search space and provide stronger evidence than asking the model to try again generally; that applies when the first implementation fails a specific assertion in continuous integration, whereas this case requires a response to a developer wants generated code to conform to existing examples.

D:

Passing behavioral tests does not guarantee a stable public interface; contract compatibility needs its own verification; it protects against the case where a patch passes tests but violates the team's public API compatibility rule, rather than the failure described here: a developer wants generated code to conform to existing examples.

E:

Explicit edge-case tests make the expected semantics observable and prevent silent incorrect defaults; this is useful for a new utility has no tests and several tricky null-handling cases, but it leaves the stated requirement to show representative local patterns rather than generic style advice unmet.

Question 156

An agentic research pipeline must propagate failures through several workers. An analysis agent reports a numeric result but its input validation failed. Which implementation decision is the best fit?

  1. Mark the calculation blocked and include the validation failure
  2. Pass a typed timeout outcome instead of a successful empty payload
  3. Return partial-result status and the identifiers of missing sources
  4. Refuse publication until mandatory worker outcomes are successful or formally waived

Explanation

Correct Answer

A:

A computed value obtained from unverified inputs should not be promoted as evidence merely because a number was produced.

Incorrect Answers

B:

Correct error semantics prevent downstream agents from mistaking unavailable data for evidence of absence; this is useful for a writing agent receives an upstream timeout labeled success, but it leaves the stated requirement to prevent use of invalid computed output unmet.

C:

This measure fits a different trigger: a search specialist returns only two of four required evidence sources. In that setting, Downstream synthesis must know that evidence collection is incomplete; treating a partial list as full coverage can distort conclusions; it does not resolve the present need to prevent use of invalid computed output.

D:

A release gate should inspect explicit prerequisite state rather than trust an optimistic synthesis paragraph; it protects against the case where the final publication stage needs all required source checks complete, rather than the failure described here: an analysis agent reports a numeric result but its input validation failed.

Question 157

An architecture review uses Claude subagents for isolated design alternatives. An agent resumes after its local repository changed since the last session. What should the architect specify in the design?

  1. Use separate worktrees or other controlled workspace isolation where changes are applied
  2. Fork the session or task state before exploring alternate designs
  3. Give the worker the deployment paths, relevant constraints, and safe tools
  4. Revalidate changed files and environment state before relying on resumed context

Explanation

Correct Answer

D:

A saved transcript is not a guarantee that disk or remote dependencies are unchanged; material state must be checked again.

Incorrect Answers

A:

Session separation by itself does not guarantee file-system isolation; independent branches need independent mutable working copies; it protects against the case where one branch makes speculative changes that should not affect the other, rather than the failure described here: an agent resumes after its local repository changed since the last session.

B:

This measure fits a different trigger: architects want to compare two approaches from the same initial repository state. In that setting, Branches can evaluate alternatives without contaminating each other's assumptions, preserving a meaningful comparison; it does not resolve the present need to refresh facts that may be stale.

C:

Scoping protects context budget and keeps the task anchored to the part of the repository it can actually assess; this is useful for a specialist needs only the deployment directory, not the full monorepo, but it leaves the stated requirement to refresh facts that may be stale unmet.

Question 158

A multi-source intelligence assistant must preserve provenance in its conclusions. A summary cites the homepage of a website but not the page containing a key claim. Which change most directly addresses this condition?

  1. Record the direct document locator and relevant passage reference
  2. Mark the source as OCR-derived and retain the image or page locator
  3. Store each value alongside publication and observation dates
  4. Label the date unknown and qualify conclusions that depend on recency

Explanation

Correct Answer

A:

A generic homepage does not let reviewers trace the asserted evidence; precise provenance makes claims auditable. For the stated case, the check is whether the system can improve citation specificity; this approach supplies that control.

Incorrect Answers

B:

Image-derived text can contain recognition errors; noting origin supports appropriate human verification; this is useful for an OCR system extracts text from a scanned image with uncertain characters, but it leaves the stated requirement to improve citation specificity unmet.

C:

This measure fits a different trigger: two sources give different values for the same metric on different dates. In that setting, Different dates may describe genuine changes rather than factual contradiction; temporal provenance enables accurate interpretation; it does not resolve the present need to improve citation specificity.

D:

A missing timestamp cannot be repaired by inference from a filename or crawl date; time-sensitive claims need verified dates; it protects against the case where a retrieved document lacks a reliable publication date, rather than the failure described here: a summary cites the homepage of a website but not the page containing a key claim.

Question 159

A code assistant must choose among repository inspection and execution tools. The team asks whether the current test suite passes after a patch. What is the most defensible next step?

  1. Invoke the approved test runner and inspect its exit status and output
  2. Allow file reads and searches while withholding file-edit tools
  3. Give the scanner read access and leave release operations to the pipeline gate
  4. Restrict edits and shell working directory to the authorized path

Explanation

Correct Answer

A:

A model-generated prediction cannot verify runtime behavior; the observed test result is the relevant evidence. This matches the workflow goal to use an actual test execution tool rather than introducing an unrelated action.

Incorrect Answers

B:

A planning phase should not silently mutate the workspace before the human approves the proposed patch; this is useful for a configuration change must be proposed but not written until review, but it leaves the stated requirement to use an actual test execution tool unmet.

C:

A worker that inspects risk should not control whether risky artifacts are published, especially before independent evaluation; it protects against the case where a scanner output must be checked for known-risk entries before publishing, rather than the failure described here: the team asks whether the current test suite passes after a patch.

D:

Worker-specific file boundaries reduce the risk that a narrowly assigned fix changes unrelated services; that applies when a test-fix worker has permission to edit only a small package, whereas this case requires a response to the team asks whether the current test suite passes after a patch.

Question 160

A document-extraction service must report the state of every processed page. A required identity field is missing from the extracted record. Which approach satisfies the stated constraint?

  1. Return an explicit missing-required-field failure
  2. Return valid page results and flag page nine for targeted correction
  3. Return expected and observed totals with a validation-failure code
  4. Report passed, failed, and needs-review counts separately
  5. Use document identity to prevent reuse of unrelated cached outputs

Explanation

Correct Answer

A:

A ledger-ready record must satisfy required business fields; a syntactically partial result cannot silently pass to commit.

Incorrect Answers

B:

This measure fits a different trigger: eight pages are parsed, but page nine contains unreadable characters. In that setting, Page-level status avoids losing successful extraction while preventing the unreadable page from being misrepresented as verified; it does not resolve the present need to block automatic posting of incomplete data.

C:

Specific diagnostics help corrective steps distinguish arithmetic conflict from parser failure or missing pages; this is useful for a semantic validator detects inconsistent tax arithmetic, but it leaves the stated requirement to block automatic posting of incomplete data unmet.

D:

Accurate status aggregation ensures that human-review items remain visible instead of being included in successful automation totals; it protects against the case where a final report says all invoices passed even though a few were escalated, rather than the failure described here: a required identity field is missing from the extracted record.

E:

Cached success for a different input is not a valid substitute; identity-bound artifacts prevent accidental data contamination; that applies when a conversion worker crashes but a coordinator has cached its last successful output from another document, whereas this case requires a response to a required identity field is missing from the extracted record.

Claude Certified Architect - Foundations (CCAR-F) (CCA-F) Practice Tests