Anthropic CCDV-F (Claude Certified Developer - Foundations) Practice Test 52

Question 510

Two separate tickets in a language-model integration backlog require different controls. One needs to call the Messages endpoint from a language without a mature provider SDK; the other needs to prevent a background processor from overwhelming the upstream API during bursts. Which implementation pair is correct?

  1. First component: Use the Messages count_tokens capability. Second component: Send the correct API authentication header.
  2. First component: Send the correct API authentication header. Second component: Honor server rate-limit signals.
  3. First component: Use direct HTTP REST integration for unsupported runtimes. Second component: Use the Messages count_tokens capability.
  4. First component: Use direct HTTP REST integration for unsupported runtimes. Second component: Honor server rate-limit signals.

Explanation

Correct Answer

D:

The first proposal fits its assigned role because a standards-based HTTP client can send JSON requests with the correct headers and decode responses, while the second proposal fits its assigned role because throttling and backoff work with provider limits rather than assuming infinite request capacity; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to prevent a background processor from overwhelming the upstream API during bursts.

Incorrect Answers

A:

The first proposal serves the different purpose to estimate model-token input demand before launching a large processing job instead of the assigned first role because token counting helps check input budgets before generation and cost planning, while the second proposal serves the different purpose to avoid unauthenticated model requests from a backend worker instead of the assigned second role because the API validates the provided credential before accepting a request; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to prevent a background processor from overwhelming the upstream API during bursts.

B:

The first proposal serves the different purpose to avoid unauthenticated model requests from a backend worker instead of the assigned first role because the API validates the provided credential before accepting a request, while the second proposal fits its assigned role because throttling and backoff work with provider limits rather than assuming infinite request capacity; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to prevent a background processor from overwhelming the upstream API during bursts.

C:

The first proposal fits its assigned role because a standards-based HTTP client can send JSON requests with the correct headers and decode responses, while the second proposal serves the different purpose to estimate model-token input demand before launching a large processing job instead of the assigned second role because token counting helps check input budgets before generation and cost planning; their respective acceptance criteria are to call the Messages endpoint from a language without a mature provider SDK and to prevent a background processor from overwhelming the upstream API during bursts.

Question 511

A production-readiness check for a workflow automation component requires the service to avoid charging a customer twice after a worker retries a timed-out payment action. Which approach passes this check?

  1. Use PreToolUse gating for dangerous operations
  2. Add idempotency keys to side-effecting operations
  3. Choose isolated sandbox execution for untrusted code
  4. Keep user-specific state out of a shared agent process

Explanation

Correct Answer

B:

Application-generated operation identifiers allow the downstream service to recognize a repeated request; this behavior directly supports the need to avoid charging a customer twice after a worker retries a timed-out payment action. Model-level instructions cannot guarantee exactly-once execution after a partial failure.

Incorrect Answers

A:

When an application needs to reject a shell command that targets protected deployment secrets before execution, a pre-execution hook can examine a proposed action and block it through a deterministic policy; the current capability must instead avoid charging a customer twice after a worker retries a timed-out payment action, while post-execution logging cannot prevent a destructive command already run.

C:

This approach serves the separate need to let an agent run generated scripts without exposing production host files because a restricted execution environment limits filesystem, network, and resource privileges; it does not establish the outcome to avoid charging a customer twice after a worker retries a timed-out payment action, and running code in the main application process expands the blast radius of a bad tool call.

D:

Per-request identity and session boundaries prevent cross-tenant context leakage; despite being useful when a service must serve many tenants without mixing their tool outputs or prior messages, it is not a solution to avoid charging a customer twice after a worker retries a timed-out payment action; shared conversation histories can inadvertently expose unrelated customers' data.

Question 512

A production API team owns two independent components. The first must decide between streaming and offline processing for a time-sensitive support assistant; the second must turn a vague request for an AI claims assistant into observable user actions and outputs. Which paired plan fulfills both?

  1. First component: Define failure and recovery objectives. Second component: Classify data sensitivity up front.
  2. First component: Record nonfunctional latency targets. Second component: Define failure and recovery objectives.
  3. First component: Identify authoritative source systems. Second component: Define failure and recovery objectives.
  4. First component: Record nonfunctional latency targets. Second component: Document functional acceptance criteria.
  5. First component: Classify data sensitivity up front. Second component: Document functional acceptance criteria.

Explanation

Correct Answer

D:

The first proposal fits its assigned role because a measurable response-time objective constrains interaction design and infrastructure choices, while the second proposal fits its assigned role because acceptance criteria specify inputs, outputs, and successful behavior that can be tested before release; their respective acceptance criteria are to decide between streaming and offline processing for a time-sensitive support assistant and to turn a vague request for an AI claims assistant into observable user actions and outputs.

Incorrect Answers

A:

The first proposal serves the different purpose to keep a regulated triage service usable during an upstream API outage instead of the assigned first role because availability, recovery time, and fallback behavior must be specified as measurable operational needs, while the second proposal serves the different purpose to decide which customer fields must be redacted before model requests instead of the assigned second role because data classification guides minimization, access controls, retention, and acceptable transfers; their respective acceptance criteria are to decide between streaming and offline processing for a time-sensitive support assistant and to turn a vague request for an AI claims assistant into observable user actions and outputs.

B:

The first proposal fits its assigned role because a measurable response-time objective constrains interaction design and infrastructure choices, while the second proposal serves the different purpose to keep a regulated triage service usable during an upstream API outage instead of the assigned second role because availability, recovery time, and fallback behavior must be specified as measurable operational needs; their respective acceptance criteria are to decide between streaming and offline processing for a time-sensitive support assistant and to turn a vague request for an AI claims assistant into observable user actions and outputs.

C:

The first proposal serves the different purpose to prevent a Claude assistant from inventing account balances when it lacks current ledger data instead of the assigned first role because source-of-truth mapping identifies which verified systems must supply facts and what freshness guarantees apply, while the second proposal serves the different purpose to keep a regulated triage service usable during an upstream API outage instead of the assigned second role because availability, recovery time, and fallback behavior must be specified as measurable operational needs; their respective acceptance criteria are to decide between streaming and offline processing for a time-sensitive support assistant and to turn a vague request for an AI claims assistant into observable user actions and outputs.

E:

The first proposal serves the different purpose to decide which customer fields must be redacted before model requests instead of the assigned first role because data classification guides minimization, access controls, retention, and acceptable transfers, while the second proposal fits its assigned role because acceptance criteria specify inputs, outputs, and successful behavior that can be tested before release; their respective acceptance criteria are to decide between streaming and offline processing for a time-sensitive support assistant and to turn a vague request for an AI claims assistant into observable user actions and outputs.

Question 513

The current design of a Claude application pipeline does not adequately stop processing when a user cancels a long report and release resources cleanly. Which improvement resolves the gap?

  1. Handle empty and partial model responses
  2. Add cancellation handling to streaming UI
  3. Treat retrieval results as untrusted data
  4. Use bounded retrieval results

Explanation

Correct Answer

B:

Cancellation signals should propagate through streams, queues, and workers where supported; this behavior directly supports the need to stop processing when a user cancels a long report and release resources cleanly. Hiding the browser output alone may leave paid requests running.

Incorrect Answers

A:

For the alternate objective to keep a downstream parser from crashing when a request returns no usable text block, content-aware response handling checks block types, stop reasons, and required values before proceeding; the feature here must stop processing when a user cancels a long report and release resources cleanly, and assuming every message contains a single complete text string is unsafe.

C:

When an application needs to prevent a malicious help article from ordering the model to reveal secrets, retrieved content should be framed as evidence, never upgraded to new tool or system authority; the current capability must instead stop processing when a user cancels a long report and release resources cleanly, while a trustworthy source URL does not make embedded commands authorized.

D:

Ranking and a context budget prioritize high-quality evidence while controlling input tokens; the supported outcome is to avoid burying the relevant contract clause beneath hundreds of low-value passages, whereas this scenario requires the ability to stop processing when a user cancels a long report and release resources cleanly; sending every possible hit can cause context bloat and distract attention.

Question 514

An architecture review of a production API product lists two distinct outcomes: the first subsystem must trace a quality regression to the exact instruction revision used, and a second subsystem must allow broad exploratory tools in a sandbox but only approved reads in production. Which pair delivers these outcomes?

  1. First component: Version prompt templates alongside code. Second component: Use separate permission profiles by environment.
  2. First component: Validate configuration at startup. Second component: Put repository guidance in CLAUDE.md.
  3. First component: Put repository guidance in CLAUDE.md. Second component: Use separate permission profiles by environment.
  4. First component: Version prompt templates alongside code. Second component: Validate configuration at startup.

Explanation

Correct Answer

A:

The first proposal fits its assigned role because source-controlled prompt assets make rollbacks, diffs, and evaluations possible, while the second proposal fits its assigned role because environment-scoped policy avoids inheriting development privileges in a live setting; their respective acceptance criteria are to trace a quality regression to the exact instruction revision used and to allow broad exploratory tools in a sandbox but only approved reads in production.

Incorrect Answers

B:

The first proposal serves the different purpose to fail fast if a model ID or tool endpoint is absent in production instead of the assigned first role because startup checks surface missing required values before user traffic reaches a broken path, while the second proposal serves the different purpose to give Claude Code the project's build, test, and architecture conventions instead of the assigned second role because repository memory files provide persistent instructional context to coding sessions; their respective acceptance criteria are to trace a quality regression to the exact instruction revision used and to allow broad exploratory tools in a sandbox but only approved reads in production.

C:

The first proposal serves the different purpose to give Claude Code the project's build, test, and architecture conventions instead of the assigned first role because repository memory files provide persistent instructional context to coding sessions, while the second proposal fits its assigned role because environment-scoped policy avoids inheriting development privileges in a live setting; their respective acceptance criteria are to trace a quality regression to the exact instruction revision used and to allow broad exploratory tools in a sandbox but only approved reads in production.

D:

The first proposal fits its assigned role because source-controlled prompt assets make rollbacks, diffs, and evaluations possible, while the second proposal serves the different purpose to fail fast if a model ID or tool endpoint is absent in production instead of the assigned second role because startup checks surface missing required values before user traffic reaches a broken path; their respective acceptance criteria are to trace a quality regression to the exact instruction revision used and to allow broad exploratory tools in a sandbox but only approved reads in production.

Question 515

A developer is reviewing alternatives for a privacy controls feature that must reduce exposure of customer documents copied into application telemetry. Which option is a defensible fit?

  1. Add an explicit export approval check
  2. Use output escaping to prevent XSS
  3. Minimize personally identifiable information
  4. Apply retention limits to diagnostic logs

Explanation

Correct Answer

D:

Data minimization and controlled retention reduce the duration and audience of sensitive log content; this behavior directly supports the need to reduce exposure of customer documents copied into application telemetry. An internal log is still a data store requiring access controls.

Incorrect Answers

A:

When an application needs to prevent unreviewed bulk downloads triggered by an agent's tool plan, a separate authorization or human approval step can gate high-impact operations; the current capability must instead reduce exposure of customer documents copied into application telemetry, while the model's confidence in its own plan does not grant export permission.

B:

Context-aware encoding prevents text from being interpreted as active HTML or JavaScript; the supported outcome is to display generated customer notes safely in a browser, whereas this scenario requires the ability to reduce exposure of customer documents copied into application telemetry; an instruction to the model not to write scripts is not a rendering control.

C:

This approach serves the separate need to avoid transmitting unnecessary customer identifiers with a summarization request because redaction and selective fields reduce exposure before data enters model or logging paths; it does not establish the outcome to reduce exposure of customer documents copied into application telemetry, and logging every full prompt can defeat otherwise careful minimization.

Question 516

An inference routing review requires two independent, explicitly defined outcomes: the first workstream must avoid confusing allowable prompt size with maximum generated response length; a second must clarify a specific formatting convention that a plain instruction was not conveying. Which TWO actions directly provide these capabilities? Choose TWO.

  1. Recognize next-token generation
  2. Use contrasting few-shot examples
  3. Use one representative example
  4. Distinguish input from output token limits
  5. Treat model outputs as stochastic

Explanation

Correct Answers

C:

A single demonstration can anchor the expected shape and tone of a generated response; this directly supports task 2 to clarify a specific formatting convention that a plain instruction was not conveying, not the separate task or tasks to avoid confusing allowable prompt size with maximum generated response length; an example is not a formal guarantee that every response will conform.

D:

The context budget constrains total material considered while output ceilings bound what a single generation can emit; this directly supports task 1 to avoid confusing allowable prompt size with maximum generated response length, not the separate task or tasks to clarify a specific formatting convention that a plain instruction was not conveying; raising an output cap does not make an oversized input fit.

Incorrect Answers

A:

The model predicts likely continuations conditioned on context rather than querying an authoritative truth database by default; this implements the different outcome to explain why a plausible answer can still contain an unsupported factual claim rather than the required tasks to avoid confusing allowable prompt size with maximum generated response length and to clarify a specific formatting convention that a plain instruction was not conveying, and fluent prose is not evidence that a source was verified, so it is not a direct solution to any of the assigned workstreams.

B:

Multiple representative cases help define a decision boundary and reduce ambiguity; this implements the different outcome to distinguish borderline policy labels with examples of both allowed and disallowed cases rather than the required tasks to avoid confusing allowable prompt size with maximum generated response length and to clarify a specific formatting convention that a plain instruction was not conveying, and repeating near-identical samples does not cover edge cases, so it is not a direct solution to any of the assigned workstreams.

E:

Non-determinism requires evaluation by criteria and repeated examples rather than exact-string equality everywhere; this implements the different outcome to design tests that tolerate benign wording variation while detecting semantic failures rather than the required tasks to avoid confusing allowable prompt size with maximum generated response length and to clarify a specific formatting convention that a plain instruction was not conveying, and a single run is insufficient to prove robust model behavior, so it is not a direct solution to any of the assigned workstreams.

Question 517

A Claude application review requires two independent, explicitly defined outcomes: the first workstream must keep a regulated triage service usable during an upstream API outage; a second must prevent a Claude assistant from inventing account balances when it lacks current ledger data. Which TWO actions directly provide these capabilities? Choose TWO.

  1. Identify authoritative source systems
  2. Define failure and recovery objectives
  3. Record nonfunctional latency targets
  4. Define input modalities and size bounds
  5. Document functional acceptance criteria

Explanation

Correct Answers

A:

Source-of-truth mapping identifies which verified systems must supply facts and what freshness guarantees apply; this directly supports task 2 to prevent a Claude assistant from inventing account balances when it lacks current ledger data, not the separate task or tasks to keep a regulated triage service usable during an upstream API outage; model inference is not a substitute for a transactional database.

B:

Availability, recovery time, and fallback behavior must be specified as measurable operational needs; this directly supports task 1 to keep a regulated triage service usable during an upstream API outage, not the separate task or tasks to prevent a Claude assistant from inventing account balances when it lacks current ledger data; a happy-path prototype misses behavior under dependency failures.

Incorrect Answers

C:

A measurable response-time objective constrains interaction design and infrastructure choices; this implements the different outcome to decide between streaming and offline processing for a time-sensitive support assistant rather than the required tasks to keep a regulated triage service usable during an upstream API outage and to prevent a Claude assistant from inventing account balances when it lacks current ledger data, and functional prompts alone cannot establish acceptable performance, so it is not a direct solution to any of the assigned workstreams.

D:

Format, resolution, size, and throughput requirements determine preprocessing and model-call strategy; this implements the different outcome to build an intake service that accepts scanned documents as well as typed text rather than the required tasks to keep a regulated triage service usable during an upstream API outage and to prevent a Claude assistant from inventing account balances when it lacks current ledger data, and assuming every input is plain text will fail at the ingestion boundary, so it is not a direct solution to any of the assigned workstreams.

E:

Acceptance criteria specify inputs, outputs, and successful behavior that can be tested before release; this implements the different outcome to turn a vague request for an AI claims assistant into observable user actions and outputs rather than the required tasks to keep a regulated triage service usable during an upstream API outage and to prevent a Claude assistant from inventing account balances when it lacks current ledger data, and choosing a model first does not define what the application must accomplish, so it is not a direct solution to any of the assigned workstreams.

Question 518

A release plan for a backend integration platform has two acceptance checks: component one must explain service limits or tool outages without leaking stack traces or secrets; component two must stop processing when a user cancels a long report and release resources cleanly. Which proposal passes both?

  1. First component: Keep user tasks in the user message. Second component: Isolate tools behind a service layer.
  2. First component: Isolate tools behind a service layer. Second component: Add cancellation handling to streaming UI.
  3. First component: Return user-friendly error states. Second component: Keep user tasks in the user message.
  4. First component: Return user-friendly error states. Second component: Add cancellation handling to streaming UI.

Explanation

Correct Answer

D:

The first proposal fits its assigned role because structured error mapping presents actionable feedback while retaining detailed diagnostics in protected logs, while the second proposal fits its assigned role because cancellation signals should propagate through streams, queues, and workers where supported; their respective acceptance criteria are to explain service limits or tool outages without leaking stack traces or secrets and to stop processing when a user cancels a long report and release resources cleanly.

Incorrect Answers

A:

The first proposal serves the different purpose to separate an end user's request from the developer's stable operating instructions instead of the assigned first role because explicit message roles distinguish requested work from higher-priority operational framing, while the second proposal serves the different purpose to avoid exposing raw production database credentials directly to an agent instead of the assigned second role because a mediation service validates identity, arguments, authorization, and audit logs for each requested action; their respective acceptance criteria are to explain service limits or tool outages without leaking stack traces or secrets and to stop processing when a user cancels a long report and release resources cleanly.

B:

The first proposal serves the different purpose to avoid exposing raw production database credentials directly to an agent instead of the assigned first role because a mediation service validates identity, arguments, authorization, and audit logs for each requested action, while the second proposal fits its assigned role because cancellation signals should propagate through streams, queues, and workers where supported; their respective acceptance criteria are to explain service limits or tool outages without leaking stack traces or secrets and to stop processing when a user cancels a long report and release resources cleanly.

C:

The first proposal fits its assigned role because structured error mapping presents actionable feedback while retaining detailed diagnostics in protected logs, while the second proposal serves the different purpose to separate an end user's request from the developer's stable operating instructions instead of the assigned second role because explicit message roles distinguish requested work from higher-priority operational framing; their respective acceptance criteria are to explain service limits or tool outages without leaking stack traces or secrets and to stop processing when a user cancels a long report and release resources cleanly.

Question 519

An architecture review of an inference routing product lists two distinct outcomes: the first subsystem must avoid selecting a model that lacks a required document or tool feature, and a second subsystem must detect behavior changes before a newly released model reaches customers. Which pair delivers these outcomes?

  1. First component: Prefer a high-capability Opus-class model. Second component: Run task-specific model evaluations.
  2. First component: Check modality and feature support. Second component: Prefer a high-capability Opus-class model.
  3. First component: Run task-specific model evaluations. Second component: Pin model versions during a migration.
  4. First component: Check modality and feature support. Second component: Pin model versions during a migration.

Explanation

Correct Answer

D:

The first proposal fits its assigned role because availability of context length, structured output, and modalities may depend on the selected model and version, while the second proposal fits its assigned role because explicit IDs and a staged rollout make differences attributable and reversible; their respective acceptance criteria are to avoid selecting a model that lacks a required document or tool feature and to detect behavior changes before a newly released model reaches customers.

Incorrect Answers

A:

The first proposal serves the different purpose to handle a rare complex architecture review where reasoning accuracy outweighs cost and latency instead of the assigned first role because the capability-oriented tier is suited to the hardest tasks after evaluation confirms an advantage, while the second proposal serves the different purpose to choose between releases using an actual organization's error and latency measurements instead of the assigned second role because representative tests expose differences hidden by generic capability labels; their respective acceptance criteria are to avoid selecting a model that lacks a required document or tool feature and to detect behavior changes before a newly released model reaches customers.

B:

The first proposal fits its assigned role because availability of context length, structured output, and modalities may depend on the selected model and version, while the second proposal serves the different purpose to handle a rare complex architecture review where reasoning accuracy outweighs cost and latency instead of the assigned second role because the capability-oriented tier is suited to the hardest tasks after evaluation confirms an advantage; their respective acceptance criteria are to avoid selecting a model that lacks a required document or tool feature and to detect behavior changes before a newly released model reaches customers.

C:

The first proposal serves the different purpose to choose between releases using an actual organization's error and latency measurements instead of the assigned first role because representative tests expose differences hidden by generic capability labels, while the second proposal fits its assigned role because explicit IDs and a staged rollout make differences attributable and reversible; their respective acceptance criteria are to avoid selecting a model that lacks a required document or tool feature and to detect behavior changes before a newly released model reaches customers.

Question 520

The current design of a retrieval pipeline pipeline does not adequately stop stale log dumps from crowding out new evidence during an agent session. Which improvement resolves the gap?

  1. Prune obsolete tool outputs
  2. Use scoped subagent contexts
  3. Refresh stale evidence before reuse
  4. Separate long-term memory from active scratchpad

Explanation

Correct Answer

A:

Tool-result pruning removes redundant bulky observations after their findings are retained; this behavior directly supports the need to stop stale log dumps from crowding out new evidence during an agent session. Keeping all raw logs forever does not guarantee better reasoning.

Incorrect Answers

B:

This approach serves the separate need to prevent a specialist's exhaustive document scan from contaminating the coordinator's working context because isolated workers return concise findings while retaining intermediate detail within their own window; it does not establish the outcome to stop stale log dumps from crowding out new evidence during an agent session, and sharing every search hit with every agent multiplies noise and cost.

C:

Freshness checks determine whether a saved passage remains authoritative for a dynamic fact; despite being useful when a service must avoid answering a pricing question from last month's retrieved cache, it is not a solution to stop stale log dumps from crowding out new evidence during an agent session; high semantic similarity says nothing about timeliness.

D:

For the alternate objective to prevent outdated user preferences from being treated as current task instructions, explicit storage and freshness rules govern what is retained and when it is reintroduced; the feature here must stop stale log dumps from crowding out new evidence during an agent session, and an unbounded persistent transcript can accumulate contradictory assumptions.

Claude Certified Developer - Foundations (CCDV-F) Practice Tests