Anthropic CCAR-P (Claude Certified Architect - Professional) Practice Test 21

Question 200

A research administration pilot cannot be accepted because long-lived broad secrets amplify compromise impact. The owner can change one primary architectural control; what should it be?

  1. Deploy Per-user delegated authorization
  2. Deploy Short-lived scoped service credentials
  3. Deploy OAuth consent and delegated scopes
  4. Deploy Authenticated webhook processing

Explanation

Correct Answer

B:

The engineering fix changes the mechanism rather than the presentation: the integration can issue narrowly scoped renewable credentials for the job identity. It requires identity provider integration and secret rotation process; a regression test should examine credential scope violations and secret age in cases where long-lived broad secrets amplify compromise impact.

Incorrect Answers

A:

When a shared service token can reveal records across users, the integration can enforce each call using the requesting user's authorized scope; but this remedy operates on another boundary, not the case where long-lived broad secrets amplify compromise impact.

C:

When administrator-wide credentials bypass user-controlled consent semantics, the integration can obtain user consent for minimum scopes and handle revocation; yet it would not correct the observed weakness: long-lived broad secrets amplify compromise impact.

D:

Although the alternative reduces the risk that trusting an inbound URL lets attackers forge operational events because the integration can verify message authenticity, timestamp and replay protection before accepting commands; but this remedy operates on another boundary, not the case where long-lived broad secrets amplify compromise impact.

Question 201

A telecommunications support audit challenges the need for Value-to-risk prioritization. Which risk would remain unmanaged if its function were not provided elsewhere?

  1. An exposure in which prioritizing solely by raw request volume ignores potential harm
  2. An exposure in which automating the last step could bypass required accountability
  3. An exposure in which deployment can appear successful while adding hidden review effort
  4. An exposure in which a polished demo can hide unacceptable long-tail failures

Explanation

Correct Answer

A:

The relevant failure mode is that prioritizing solely by raw request volume ignores potential harm. The existing design can prevent it when a scored decision framework surfaces high-value lower-risk starting points. The team needs a joint rubric agreed with security and operations and must monitor risk-adjusted impact and implementation effort per use case to support any claim of equivalent protection.

Incorrect Answers

B:

The competing hazard is that automating the last step could bypass required accountability in a workload where caseworkers must retain authority for consequential final decisions; whereas removing Value-to-risk prioritization primarily exposes the risk that prioritizing solely by raw request volume ignores potential harm.

C:

When executives claim the assistant will save time without measuring current work, deployment can appear successful while adding hidden review effort; but this is not the immediate loss associated with omitting Value-to-risk prioritization: prioritizing solely by raw request volume ignores potential harm.

D:

The competing hazard is that a polished demo can hide unacceptable long-tail failures in a workload where the proposed automation encounters exceptions that have never been tested; however, the named safeguard addresses the distinct hazard that prioritizing solely by raw request volume ignores potential harm.

Question 202

A compliance assurance rollout must satisfy this design premise: the integration can issue narrowly scoped renewable credentials for the job identity. Choose TWO linked items that form a workable architecture change and its essential preparation. Choose TWO.

  1. Track blocked unauthorized operations in adversarial tool-call tests
  2. Establish propagated identity and resource-level access policy
  3. Establish identity provider integration and secret rotation process
  4. Implement Short-lived scoped service credentials
  5. Implement OAuth consent and delegated scopes

Explanation

Correct Answers

C:

This step creates the enabling basis of identity provider integration and secret rotation process. That preparation supports the mechanism that the integration can issue narrowly scoped renewable credentials for the job identity; merely choosing the control does not prove its configuration or authority is ready.

D:

The architecture component is justified by its behavior: the integration can issue narrowly scoped renewable credentials for the job identity. It targets the exposure that long-lived broad secrets amplify compromise impact, while the complementary selected actions provide readiness and, for a three-part package, evidence.

Incorrect Answers

A:

This test would validate a different benefit: the executor evaluates policy before invoking protected downstream APIs; but that indicator does not verify credential scope violations and secret age for the selected architecture.

B:

The alternative setup enables a different mechanism: the integration can enforce each call using the requesting user's authorized scope; whereas this intervention requires identity provider integration and secret rotation process as its operating precondition.

E:

This design choice is aimed at cases where employees connect personal enterprise resources through an integration; it is effective because the integration can obtain user consent for minimum scopes and handle revocation. The rollout described is instead governed by the case where unattended background jobs need limited read access to a narrow dataset and requires a different primary intervention; however, the intended intervention requires that the integration can issue narrowly scoped renewable credentials for the job identity.

Question 203

A compliance assurance rollout must satisfy this design premise: the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels. Choose TWO linked items that form a workable architecture change and its essential preparation. Choose TWO.

  1. Implement Security policy violation rate
  2. Establish complete cost attribution and ticket acceptance rules
  3. Establish gold labels including unanswerable requests
  4. Track verified factual claims divided by audited claims
  5. Implement Calibration and abstention quality

Explanation

Correct Answers

C:

Before rollout, the team needs gold labels including unanswerable requests. This is a deployment dependency for the intervention whose effect is that the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels, not an alternative metric or unrelated design feature.

E:

This is the design control: the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels. It addresses the hazard that accuracy on answered cases alone can reward refusing everything; the other selected element supplies the deployment foundation and, where required, the measurement of that control.

Incorrect Answers

A:

This design choice is aimed at cases where agent integration must respect access restrictions for multiple tenants; however, the intended intervention requires that the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels.

B:

The alternative setup enables a different mechanism: the evaluation process can include review, retries and inference in outcome-based unit economics; whereas this intervention requires gold labels including unanswerable requests as its operating precondition.

D:

This test would validate a different benefit: the evaluation process can score whether factual claims are supported by cited source spans; but that indicator does not verify risk-coverage curve and improper certainty rate for the selected architecture.

Question 204

The healthcare operations staging review finds that a strong model cannot compensate for absent or unauthorized source data. Which design intervention is the most appropriate corrective action?

  1. Apply Data readiness assessment
  2. Apply Current-state service blueprint
  3. Apply Workflow observation interviews
  4. Apply Use-case prioritization workshop
  5. Apply Acceptance criteria negotiation
  6. Apply Stakeholder and decision-rights mapping

Explanation

Correct Answer

A:

The remediation acts at the relevant design boundary: the delivery team can validate quality, permissions, update cadence and ownership before promising RAG results. That counteracts the observed hazard that a strong model cannot compensate for absent or unauthorized source data. Rollout should include accessible data catalog and source stewards, with regression evidence from source coverage and freshness pass rate on sampled queries.

Incorrect Answers

B:

This addresses the failure where redesigning a guessed process can shift work rather than remove it, because the delivery team can chart the existing input-to-outcome process and exception volumes; yet it would not correct the observed weakness: a strong model cannot compensate for absent or unauthorized source data.

C:

Although the alternative reduces the risk that executive summaries alone omit edge cases that determine project success because the delivery team can observe actual exception handling and frontline work before mapping automation; but this remedy operates on another boundary, not the case where a strong model cannot compensate for absent or unauthorized source data.

D:

This addresses the failure where political priority alone may select a low-return high-risk deployment, because the delivery team can jointly rank candidates by value, feasibility, harm and data readiness; whereas the healthcare operations defect follows the path in which a strong model cannot compensate for absent or unauthorized source data.

E:

Where stakeholders ask for a system that is accurate enough without specifying thresholds, the delivery team can turn goals into testable task-specific, risk-sensitive success criteria; yet it would not correct the observed weakness: a strong model cannot compensate for absent or unauthorized source data.

F:

Where product, security and legal teams disagree on who can approve release, the delivery team can define accountable decision owners and consultation rights early; but this remedy operates on another boundary, not the case where a strong model cannot compensate for absent or unauthorized source data.

Question 205

An independent assessment of a research administration Claude system reports that instruction blending permits external authors to steer agent behavior. Which change best closes the exposed control gap?

  1. Deploy High-risk human approval gate
  2. Deploy Untrusted retrieved-content isolation
  3. Deploy Ingress content classification
  4. Deploy Rate and spend limits by tenant
  5. Deploy Defense-in-depth incident response

Explanation

Correct Answer

B:

This intervention addresses the failure at its source because the control layer can treat retrieved text as evidence and prohibit it from changing tool policy. To keep the issue from recurring, establish parsing boundary and policy enforcement outside the model and verify policy override rate in hostile-document evaluations; otherwise instruction blending permits external authors to steer agent behavior can persist.

Incorrect Answers

A:

Where proposed external transfers could produce irreversible customer harm, the control layer can stop before execution and require authorized human confirmation; yet it would not correct the observed weakness: instruction blending permits external authors to steer agent behavior.

C:

The alternative reduces the risk that prompt-only guidance cannot neutralize malicious external text because the control layer can classify untrusted inputs and prevent them from acquiring instruction authority; however, the research administration defect follows the path in which instruction blending permits external authors to steer agent behavior.

D:

Although the alternative reduces the risk that a global cost cap allows one tenant to crowd out others because the control layer can enforce bounded usage budgets and anomaly controls per tenant; yet it would not correct the observed weakness: instruction blending permits external authors to steer agent behavior.

E:

When a single classifier cannot handle every emergent attack pattern, the control layer can combine prevention, detection, containment and audited recovery; but this remedy operates on another boundary, not the case where instruction blending permits external authors to steer agent behavior.

Question 206

A human-resources services team is advancing Per-class model policy beyond its proof of concept. Engineers expect that the model configuration can set explicit allowed model configurations by data class and purpose. What evidence of setup readiness should be required first?

  1. Prepare benchmark cases with multi-hop evidence
  2. Prepare measured task complexity and escalation thresholds
  3. Prepare model contract compatibility tests and outage policy
  4. Prepare approved data classification and deployment policy

Explanation

Correct Answer

D:

The prerequisite creates the necessary operating boundary: approved data classification and deployment policy. The system is intended to work because the model configuration can set explicit allowed model configurations by data class and purpose. A release without that foundation may still suffer a single default model policy may violate different governance requirements; use policy-compliant invocation rate by workload class to evaluate the result.

Incorrect Answers

A:

This prerequisite supports a method used when multi-document arbitration requires several interdependent inferences, because a stronger model can justify tradeoffs across heterogeneous evidence; but the production dependency for Per-class model policy is approved data classification and deployment policy.

B:

This prerequisite supports a method used when most messages are short and low-complexity while a minority need deep analysis, because a fast adequate tier handles easy cases and escalates only uncertain ones; for this release, the team must first verify approved data classification and deployment policy.

C:

For the separate challenge where primary inference service can time out during a critical support window, the model configuration can route to an approved fallback model with compatible output contracts; however, the named intervention needs approved data classification and deployment policy before its protection can be enforced.

Question 207

A logistics coordination service is moving to production with this aim: the engineering workflow can pin dependencies, tooling and environment assumptions across contributors. Choose TWO complementary choices, an architectural control and the precondition it depends on. Choose TWO.

  1. Establish versioned setup scripts and build reproducibility checks
  2. Establish team permission policy and escalation route
  3. Implement Hook-based policy checks
  4. Track standards adherence across fresh developer workspaces
  5. Implement Reproducible development environment

Explanation

Correct Answers

A:

The prerequisite is versioned setup scripts and build reproducibility checks. It provides the operating preparation for a system in which the engineering workflow can pin dependencies, tooling and environment assumptions across contributors; without this setup, the planned safeguard has not been demonstrated as deployable.

E:

This is the design control: the engineering workflow can pin dependencies, tooling and environment assumptions across contributors. It addresses the hazard that ambient local settings conceal incompatibilities in shared builds; the other selected element supplies the deployment foundation and, where required, the measurement of that control.

Incorrect Answers

B:

The alternative setup enables a different mechanism: the engineering workflow can configure approved actions and deny risky operations by default; whereas this intervention requires versioned setup scripts and build reproducibility checks as its operating precondition.

C:

This design choice is aimed at cases where code generation sometimes bypasses mandatory lint and secret scans; it is effective because the engineering workflow can run deterministic checks at controlled workflow boundaries. The rollout described is instead governed by the case where agent-assisted fixes work on one laptop but fail in CI and requires a different primary intervention; however, the intended intervention requires that the engineering workflow can pin dependencies, tooling and environment assumptions across contributors.

D:

This test would validate a different benefit: the engineering workflow can version shared Claude Code guidance alongside the project and review changes; but that indicator does not verify agent-generated changes passing clean-environment builds for the selected architecture.

Question 208

In an internal audit modernization project, a model can propose actions but must not decide its own privileges. Which approach best fits this specific operating constraint?

  1. Introduce OAuth consent and delegated scopes
  2. Introduce Authenticated webhook processing
  3. Introduce Server-side tool authorization
  4. Introduce Short-lived scoped service credentials

Explanation

Correct Answer

C:

The mechanism is specific: the executor evaluates policy before invoking protected downstream APIs. A credible implementation requires trusted auth context and policy engine, while its acceptance test monitors blocked unauthorized operations in adversarial tool-call tests. The alternative approaches leave the risk that prompt rules alone cannot serve as a security boundary.

Incorrect Answers

A:

If employees connect personal enterprise resources through an integration, the integration can obtain user consent for minimum scopes and handle revocation; it does not ensure the required behavior: the executor evaluates policy before invoking protected downstream APIs.

B:

If an upstream application posts actions to a model-driven workflow, the integration can verify message authenticity, timestamp and replay protection before accepting commands; whereas the supply-chain planning case must guard against the outcome that prompt rules alone cannot serve as a security boundary.

D:

For a workload where unattended background jobs need limited read access to a narrow dataset, the integration can issue narrowly scoped renewable credentials for the job identity; it does not ensure the required behavior: the executor evaluates policy before invoking protected downstream APIs.

Question 209

A supply-chain planning operational excellence team is designing a Claude-supported workflow because incoming tickets have missing identifiers and mixed attachment types. Which architectural decision most directly meets that requirement?

  1. Implement Preprocessing and sensitive-field minimization
  2. Implement Feedback taxonomy with ground-truth capture
  3. Implement Idempotent action boundary
  4. Implement Audit-correlated processing stages
  5. Implement Validated input contract

Explanation

Correct Answer

E:

The proposed control fits because schema checks and normalization reject malformed inputs before they enter the model context. Its production prerequisites include an explicit schema and ownership of input validation; check invalid-input rate and successful recovery by input category on representative tasks. Without this boundary, unvalidated inputs propagate ambiguity into generated actions.

Incorrect Answers

A:

When raw customer messages contain identifiers irrelevant to classification, preprocessing redacts unnecessary fields while preserving task-bearing content; whereas the supply-chain planning case must guard against the outcome that unvalidated inputs propagate ambiguity into generated actions.

B:

If reviewers only click thumbs-up or thumbs-down without reason codes, structured error tags make future fixes attributable to retrieval, prompt or policy causes; the stated scenario depends on a different mechanism: schema checks and normalization reject malformed inputs before they enter the model context.

C:

When a retrying agent sometimes creates the same support task twice, idempotency keys and deduplication protect side-effecting writes across retries; that mechanism leaves another need unresolved: schema checks and normalization reject malformed inputs before they enter the model context.

D:

When compliance teams cannot reconstruct which tool outputs informed a recommendation, stage-level correlation links input versions, retrieval results, model calls and final decisions; whereas the supply-chain planning case must guard against the outcome that unvalidated inputs propagate ambiguity into generated actions.

Question 210

During solution discovery in procurement services, long-tail failure classes rarely appear in a random convenience sample. Which design response addresses the stated constraint most directly?

  1. Deploy Held-out evaluation split
  2. Deploy Stratified production-like test set
  3. Deploy Automated contract test suite
  4. Deploy Temporal holdout for changing policies

Explanation

Correct Answer

B:

The intervention acts on the system behavior directly: the evaluation process can stratify by task type, risk, language and source complexity. Before rollout, verify privacy-cleared examples and a sampling design. The relevant success test is subgroup coverage and performance against known strata; otherwise a high average score can mask catastrophic minority failures.

Incorrect Answers

A:

When prompt engineers tune repeatedly on the same example corpus, the evaluation process can reserve untouched cases for final comparison after iteration; that does not provide the capability actually needed: the evaluation process can stratify by task type, risk, language and source complexity.

C:

When downstream software expects exact structured fields and allowed enum values, the evaluation process can validate outputs mechanically on every release candidate; whereas the procurement services case must guard against the outcome that a high average score can mask catastrophic minority failures.

D:

When policy retrieval quality looks good using documents from the test period, the evaluation process can test on later source changes to measure freshness and drift response; that does not provide the capability actually needed: the evaluation process can stratify by task type, risk, language and source complexity.

Claude Certified Architect - Professional (CCAR-P) Practice Tests