Anthropic CCA-F (Claude Certified Architect - Foundations (CCAR-F)) Practice Test 5

Question 40

A document classifier uses Claude to tag contracts with legal categories. A particular clause belongs to two related categories but reviewers expect one primary label. What should the architect specify in the design?

  1. Replace obsolete examples and rerun evaluation on current cases
  2. Include labeled examples that explain how to break ties
  3. Show benign text that resembles the target but does not satisfy its criteria
  4. Choose examples from different layouts and writing styles while keeping the rubric stable
  5. Select diverse boundary cases based on observed error patterns

Explanation

Correct Answer

B:

Examples should show which category wins at overlapping boundaries rather than leaving Claude to infer an unstated hierarchy.

Incorrect Answers

A:

Few-shot content can encode policy mistakes as strongly as explicit instructions; stale exemplars should be treated as configuration debt; the alternative problem occurs when a prompt example contains outdated policy thresholds. The present scenario instead requires the system to demonstrate category precedence.

C:

It would be appropriate when reviewers identify recurring false positives from harmless boilerplate; negative cases calibrate the decision boundary more directly than adding more obvious positive examples. Here, a particular clause belongs to two related categories but reviewers expect one primary label, so a different mechanism is required.

D:

Varied examples prevent the model from associating a label with incidental formatting instead of the clause meaning; it protects against the case where the training examples use only one document style, but production inputs vary, rather than the failure described here: a particular clause belongs to two related categories but reviewers expect one primary label.

E:

This choice is relevant when the team has hundreds of redundant examples in a single prompt; a smaller curated set often gives clearer task guidance and avoids burying the decision rule under duplicated demonstrations. Here, however, a particular clause belongs to two related categories but reviewers expect one primary label.

Question 41

An invoice-extraction team builds reliable prompting for standardized records. Extracted dates sometimes use locale-dependent shorthand. Which change most directly addresses this condition?

  1. Require a null or explicit missing status instead of an invented identifier
  2. Define total, subtotal, tax, and discount semantics with examples
  3. State the target ISO date format and how uncertain dates must be handled
  4. Require page or span references for important extracted fields

Explanation

Correct Answer

C:

Formatting criteria prevent ambiguous month/day interpretations and make downstream type checks meaningful. This matches the workflow goal to define the required date representation rather than introducing an unrelated action.

Incorrect Answers

A:

It would be appropriate when a document is missing a purchase-order number; missing fields should remain missing in the structured output; completeness must not be achieved by hallucinating values. Here, extracted dates sometimes use locale-dependent shorthand, so a different mechanism is required.

B:

Financial field names need precise meanings; otherwise valid-looking numbers can be assigned to the wrong ledger fields; it protects against the case where the prompt says 'extract the total' without distinguishing gross and net amounts, rather than the failure described here: extracted dates sometimes use locale-dependent shorthand.

D:

Evidence-linked extraction allows downstream review to inspect the actual text behind uncertain or disputed values; the alternative problem occurs when a reviewer needs a confidence explanation tied to source text. The present scenario instead requires the system to define the required date representation.

Question 42

An internal compliance system tracks where evidence for decisions originated. A policy interpretation quotes a paragraph but does not identify the policy revision. What is the most defensible next step?

  1. Carry source ID, page, and extraction run identifier into the final record
  2. Mark quotations, extracted facts, and analyst inferences separately
  3. Compare source-of-truth ownership and freshness before selecting a status
  4. Attach the document version and effective date to the quoted passage

Explanation

Correct Answer

D:

Policy text changes over time; revision identity prevents a once-valid rule from being applied to the wrong period.

Incorrect Answers

A:

It would be appropriate when an extraction result is copied into a report with no link to its document; a provenance chain connects the published value to the observed document and the specific processing run that produced it. Here, a policy interpretation quotes a paragraph but does not identify the policy revision, so a different mechanism is required.

B:

Separating evidence types prevents a model's synthesis from appearing to be a verbatim statement by an external authority; the alternative problem occurs when a research report mixes direct quotations and model paraphrases without labeling them. The present scenario instead requires the system to make policy claims revision-specific.

C:

A newer timestamp is not always sufficient if one system is a derived cache; authority and synchronization matter; it protects against the case where two tools return different customer statuses from separate data stores, rather than the failure described here: a policy interpretation quotes a paragraph but does not identify the policy revision.

Question 43

A research writing assistant must deliver claims that can be independently checked. A generated comparison uses qualitative adjectives with no supporting figures. Which approach satisfies the stated constraint?

  1. Define target sections, approximate length, and mandatory findings
  2. Require numeric support or qualified sourcing for comparative claims
  3. List approved source identifiers and prohibit unsupported external assertions
  4. Require effective dates and labels for superseded rules

Explanation

Correct Answer

B:

A clear evaluation rule distinguishes evidence-based comparison from untestable promotional language. Here it serves the explicit need to make the evidentiary standard measurable without adding a separate unverified inference.

Incorrect Answers

A:

A measurable structure improves consistency without requiring every word to follow a rigid template; it protects against the case where the task asks for a concise brief but never sets a length or section requirement, rather than the failure described here: a generated comparison uses qualitative adjectives with no supporting figures.

C:

It would be appropriate when the writer may use only materials approved by the research team; a bounded source contract limits answer drift and gives reviewers a concrete basis for rejecting unsupported material. Here, a generated comparison uses qualitative adjectives with no supporting figures, so a different mechanism is required.

D:

Time qualifiers are part of factual correctness when documents describe changing systems; bare claims can mislead; the alternative problem occurs when a report must distinguish current policies from historical descriptions. The present scenario instead requires the system to make the evidentiary standard measurable.

Question 44

A document-extraction service must report the state of every processed page. Eight pages are parsed, but page nine contains unreadable characters. Which mechanism should be selected?

  1. Return an explicit missing-required-field failure
  2. Return valid page results and flag page nine for targeted correction
  3. Report passed, failed, and needs-review counts separately
  4. Return expected and observed totals with a validation-failure code

Explanation

Correct Answer

B:

Page-level status avoids losing successful extraction while preventing the unreadable page from being misrepresented as verified.

Incorrect Answers

A:

A ledger-ready record must satisfy required business fields; a syntactically partial result cannot silently pass to commit; it protects against the case where a required identity field is missing from the extracted record, rather than the failure described here: eight pages are parsed, but page nine contains unreadable characters.

C:

This choice is relevant when a final report says all invoices passed even though a few were escalated; accurate status aggregation ensures that human-review items remain visible instead of being included in successful automation totals. Here, however, eight pages are parsed, but page nine contains unreadable characters.

D:

It would be appropriate when a semantic validator detects inconsistent tax arithmetic; specific diagnostics help corrective steps distinguish arithmetic conflict from parser failure or missing pages. Here, eight pages are parsed, but page nine contains unreadable characters, so a different mechanism is required.

Question 45

A coordinator divides a codebase-migration assessment among specialist agents. One worker is responsible for dependency scanning and another for test-risk analysis. What should be implemented at the appropriate boundary?

  1. Send only relevant file maps and task-specific excerpts to each specialist
  2. Define distinct tasks and structured outputs for each worker
  3. Return a structured blocker status with severity and supporting evidence
  4. Have the coordinator evaluate worker evidence and generate the consolidated recommendation
  5. Wait for both prerequisite worker results before launching the integration-check task

Explanation

Correct Answer

B:

Explicit outputs make the coordinator's merge traceable and prevent the workers from competing over the same incomplete artifact.

Incorrect Answers

A:

Broadcasting the whole repository inflates context and cost and obscures what each worker must evaluate; the alternative problem occurs when the coordinator can pass a 200-page repository dump to every worker. The present scenario instead requires the system to keep specialist deliverables separable.

C:

It would be appropriate when a security worker discovers an immediate blocker during parallel migration analysis; a typed blocker can halt or reroute downstream steps; a buried sentence is easy for a later summarizer to miss. Here, one worker is responsible for dependency scanning and another for test-risk analysis, so a different mechanism is required.

D:

Specialists provide evidence and domain judgments; synthesis ownership prevents conflicting independently published conclusions; it protects against the case where the team wants a single conclusion after three worker assessments, rather than the failure described here: one worker is responsible for dependency scanning and another for test-risk analysis.

E:

This choice is relevant when an integration test needs both migration and dependency analyses to finish; a join condition prevents testing a proposed change against an incomplete dependency or migration plan. Here, however, one worker is responsible for dependency scanning and another for test-risk analysis.

Question 46

An enterprise support assistant uses lengthy conversation and policy histories. The governing refund rule appears only once in the middle of a very long context. Which option best meets the operational need?

  1. Treat retrieved text as evidence, not an override of higher-priority controls
  2. Restate a concise current policy excerpt near the final task instructions
  3. Maintain a versioned decision ledger with approval status and source-turn references
  4. Fetch the latest verified address and label old entries as historical

Explanation

Correct Answer

B:

Important details can be overlooked in long contexts; placing the controlling rule near use reduces lost-in-the-middle errors.

Incorrect Answers

A:

External material can be adversarial or incorrect; long context must not promote a source document into policy authority; the alternative problem occurs when a retrieved source conflicts with an earlier system-defined safety requirement. The present scenario instead requires the system to make critical rules salient near the decision.

C:

It would be appropriate when the support chat contains several competing ticket resolutions, including proposals that were later rejected; a decision ledger tracks which resolution was authorized and which proposals were reversed, preventing a condensed conversation from treating tentative advice as binding. Here, the governing refund rule appears only once in the middle of a very long context, so a different mechanism is required.

D:

Chronological conversation content can contain stale facts; an authoritative lookup prevents outdated messages from driving action; it protects against the case where a chat history includes several superseded account addresses, rather than the failure described here: the governing refund rule appears only once in the middle of a very long context.

Question 47

An architecture review uses Claude subagents for isolated design alternatives. Architects want to compare two approaches from the same initial repository state. What is the most appropriate design choice?

  1. Give the worker the deployment paths, relevant constraints, and safe tools
  2. Revalidate changed files and environment state before relying on resumed context
  3. Record the branch identifier, assumptions, and tested outcomes for each alternative
  4. Fork the session or task state before exploring alternate designs

Explanation

Correct Answer

D:

Branches can evaluate alternatives without contaminating each other's assumptions, preserving a meaningful comparison.

Incorrect Answers

A:

It would be appropriate when a specialist needs only the deployment directory, not the full monorepo; scoping protects context budget and keeps the task anchored to the part of the repository it can actually assess. Here, architects want to compare two approaches from the same initial repository state, so a different mechanism is required.

B:

A saved transcript is not a guarantee that disk or remote dependencies are unchanged; material state must be checked again; it protects against the case where an agent resumes after its local repository changed since the last session, rather than the failure described here: architects want to compare two approaches from the same initial repository state.

C:

Explicit branch history makes comparisons auditable; merged prose can otherwise conceal which experiment produced each result; the alternative problem occurs when the team needs to reproduce why two design branches diverged. The present scenario instead requires the system to fork independent analyses from a common baseline.

Question 48

A Claude Agent SDK workflow delegates repository investigations to subagents. A worker should be able to inspect code but not run destructive shell commands. Which response is most reliable?

  1. Give the Task-oriented specialist a concrete goal, scope, and expected handoff
  2. Restrict the worker's allowedTools to safe inspection operations
  3. Run the investigation in a subagent and return a concise evidence-based summary
  4. Have the worker report changed or inspected paths and evidence for its conclusion

Explanation

Correct Answer

B:

Tool allowlists define an execution boundary that remains effective even when the worker's text suggests a dangerous action.

Incorrect Answers

A:

Structured delegated tasks clarify which worker owns the operation and how its outcome will return to the parent; the alternative problem occurs when a parent agent needs a specialist to take a bounded action rather than just offer advice. The present scenario instead requires the system to constrain a subagent's execution capabilities.

C:

An isolated worker context avoids flooding the coordinator with irrelevant file reads while preserving the actionable findings; it protects against the case where a specialist's large file-search transcript is not useful to the parent, rather than the failure described here: a worker should be able to inspect code but not run destructive shell commands.

D:

This choice is relevant when a subagent returns a broad recommendation without citing inspected files; the parent can validate a specialist's claims only when the output identifies what was examined or modified. Here, however, a worker should be able to inspect code but not run destructive shell commands.

Question 49

An invoice-extraction workflow must recover from malformed output. A tool payload is not valid against the required JSON Schema. What should the team do to resolve this particular gap?

  1. Stop after the configured retry limit and return a review-needed state
  2. Require evidence or field-level source matching for supplier identity
  3. Report the mismatched field and constraint, then retry within a bounded policy
  4. Run business-rule arithmetic validation and request targeted correction

Explanation

Correct Answer

C:

Specific schema errors give the model a repair target; blindly regenerating without feedback repeats the same failure mode.

Incorrect Answers

A:

Unlimited repair attempts can waste resources and may still fabricate a plausible record; bounded retries preserve reliability; the alternative problem occurs when the model has failed the same correction twice. The present scenario instead requires the system to return precise structural validation feedback.

B:

This choice is relevant when a corrected output has valid JSON but an invented supplier name; syntax repair does not imply factual correctness; source-backed validation checks whether each critical value is actually present. Here, however, a tool payload is not valid against the required JSON Schema.

D:

A valid numeric type does not prove accounting consistency; arithmetic rules catch errors outside the schema's expressiveness; it protects against the case where an extracted total passes type validation but does not equal subtotal plus taxes, rather than the failure described here: a tool payload is not valid against the required JSON Schema.

Question 50

Claude Code is working with an unfamiliar repository too large to read at once. An agent cannot tell where major services live in a large monorepo. Which control is most directly relevant?

  1. Store the current hypothesis, inspected files, pending test, and next action
  2. Merge file-indexed findings with status and unresolved issues
  3. Read the two implicated files and retain references to the other hits
  4. Record directories, ownership boundaries, and key entry points before deep reading
  5. Retain constraints, decisions, evidence references, and remaining risks

Explanation

Correct Answer

D:

A file map supports targeted exploration and prevents repeatedly flooding context with unrelated source files. The relevant requirement is to create a concise navigation map, and this measure operates at that decision point.

Incorrect Answers

A:

A resumable scratchpad prevents rediscovery and reduces the risk of treating an old hypothesis as a verified fix; the alternative problem occurs when a lengthy investigation is paused before the final patch. The present scenario instead requires the system to create a concise navigation map.

B:

It would be appropriate when several subagents have inspected separate subsystems; a structured handoff preserves useful specialist evidence while limiting the coordinator's context load. Here, an agent cannot tell where major services live in a large monorepo, so a different mechanism is required.

C:

Focused context makes the debugging task tractable without deleting the ability to revisit excluded evidence; it protects against the case where search results span many modules but only two files are relevant to the current bug, rather than the failure described here: an agent cannot tell where major services live in a large monorepo.

E:

This choice is relevant when the working context is compacted after many exploratory reads; useful compaction shortens history while keeping the information needed to continue safely and verify the eventual result. Here, however, an agent cannot tell where major services live in a large monorepo.

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