Anthropic CCDV-F (Claude Certified Developer - Foundations) Practice Test 78

Question 770

A design review for a high-volume Messages API gateway asks for support to support Claude models accessed through different supported cloud offerings, together with a way to handle a safety refusal as a distinct application outcome. Select the two appropriate changes. Choose TWO.

  1. Detect transport interruption and mark incomplete generated content.
  2. Configure an available server-side tool with the required API definition.
  3. Detect refusal and apply the product's allowed fallback or messaging policy.
  4. Use the provider-specific endpoint, authentication, and request conventions.
  5. Include a typed image content block in the message.

Explanation

Correct Answers

C:

To handle a safety refusal as a distinct application outcome as part of a design that must also support Claude models accessed through different supported cloud offerings, a refusal is not the same as a network failure; blindly retrying identical input may not produce an appropriate user experience; this is why the action is one of the two required selections.

D:

To support Claude models accessed through different supported cloud offerings as part of a design that must also handle a safety refusal as a distinct application outcome, Claude API, Bedrock, and Vertex-style integrations are not interchangeable URLs; host authentication and service wrappers differ; this is why the action is one of the two required selections.

Incorrect Answers

A:

This alternative serves the separate goal to avoid silently treating a broken streaming connection as a complete answer because network termination before a final message event can leave a partial response, which needs explicit recovery rather than success handling; the required outcomes are instead to support Claude models accessed through different supported cloud offerings and to handle a safety refusal as a distinct application outcome.

B:

This alternative serves the separate goal to use provider-hosted search or execution without implementing the tool body locally because provider-executed tools are handled in Anthropic's environment and differ from client functions that the caller must run; the required outcomes are instead to support Claude models accessed through different supported cloud offerings and to handle a safety refusal as a distinct application outcome.

E:

This alternative serves the separate goal to ask Claude to interpret a supported image supplied as part of a request because multimodal input must use the API's image representation and supported sourcing; merely writing a filename in text does not transfer pixels; the required outcomes are instead to support Claude models accessed through different supported cloud offerings and to handle a safety refusal as a distinct application outcome.

Question 771

A production readiness review of a hosted document-processing service finds that the team cannot yet display partial sentences without publishing malformed structured records or prevent a document-upload endpoint from passing unsupported bytes to the model. Which paired changes directly remedy the two gaps?

  1. Perform both steps: (1) Buffer structured data until it can be parsed and validated before UI commit; (2) apply content checks at suitable input and output boundaries.
  2. Perform both steps: (1) Persist explicit job states with allowed transitions; (2) validate file type, size, and allowed content conversion before api submission.
  3. Perform both steps: (1) Buffer structured data until it can be parsed and validated before UI commit; (2) validate file type, size, and allowed content conversion before api submission.
  4. Perform both steps: (1) Persist explicit job states with allowed transitions; (2) apply content checks at suitable input and output boundaries.

Explanation

Correct Answer

C:

The first change enables the team to display partial sentences without publishing malformed structured records because a text stream can arrive in fragments; consumers must distinguish presentation updates from complete machine-readable objects; the second addresses the separate need to prevent a document-upload endpoint from passing unsupported bytes to the model because client filenames are not enough to establish content validity; ingestion must enforce protocol and product constraints.

Incorrect Answers

A:

The first step is appropriate for the requirement to display partial sentences without publishing malformed structured records because a text stream can arrive in fragments; consumers must distinguish presentation updates from complete machine-readable objects; the other step instead aims to filter risky user-visible content according to application policy because moderation is a defense layer applied around generation, not a substitute for authenticated tool access or scoped retrieval, which does not meet the second requirement to prevent a document-upload endpoint from passing unsupported bytes to the model.

B:

The second step meets the requirement to prevent a document-upload endpoint from passing unsupported bytes to the model because client filenames are not enough to establish content validity; ingestion must enforce protocol and product constraints; the first step would only help to distinguish queued, running, waiting-for-approval, failed, and completed agent jobs because a state machine prevents an incomplete tool operation from being shown as successful and makes retries and cancellation predictable, leaving the separate requirement to display partial sentences without publishing malformed structured records unmet.

D:

Both steps target different results—distinguish queued, running, waiting-for-approval, failed, and completed agent jobs and filter risky user-visible content according to application policy—because a state machine prevents an incomplete tool operation from being shown as successful and makes retries and cancellation predictable and moderation is a defense layer applied around generation, not a substitute for authenticated tool access or scoped retrieval; the required capabilities are instead to display partial sentences without publishing malformed structured records and to prevent a document-upload endpoint from passing unsupported bytes to the model.

Question 772

An architecture decision for a TypeScript Claude integration must support catch unsafe credentials and incorrect tool calls before deployment and avoid retrying a malformed request that the API consistently rejects. Which option gives a complete rather than partial solution?

  1. Perform both steps: (1) Require peer review focused on API, security, and tool-side effects; (2) reuse an appropriately configured http client or sdk connection pool.
  2. Perform both steps: (1) Use strict tool input validation for the exposed tool interface; (2) classify permanent client errors separately from transient failures.
  3. Perform both steps: (1) Require peer review focused on API, security, and tool-side effects; (2) classify permanent client errors separately from transient failures.
  4. Perform both steps: (1) Use strict tool input validation for the exposed tool interface; (2) reuse an appropriately configured http client or sdk connection pool.

Explanation

Correct Answer

C:

The first change enables the team to catch unsafe credentials and incorrect tool calls before deployment because human review should check trust boundaries and external actions in addition to style; successful compilation cannot prove these behaviors are safe; the second addresses the separate need to avoid retrying a malformed request that the API consistently rejects because a deterministic validation or authorization error requires fixing the request or credential, not another identical attempt.

Incorrect Answers

A:

The first step is appropriate for the requirement to catch unsafe credentials and incorrect tool calls before deployment because human review should check trust boundaries and external actions in addition to style; successful compilation cannot prove these behaviors are safe; the other step instead aims to reduce repeated TLS setup overhead for high-volume application traffic because connection reuse lowers transport overhead but still requires limits and health handling for long-lived clients, which does not meet the second requirement to avoid retrying a malformed request that the API consistently rejects.

B:

The second step meets the requirement to avoid retrying a malformed request that the API consistently rejects because a deterministic validation or authorization error requires fixing the request or credential, not another identical attempt; the first step would only help to reject tool arguments that fail a declared shape before tool execution because a strict schema constrains the arguments a model may emit, but application authorization and business-rule checks remain necessary, leaving the separate requirement to catch unsafe credentials and incorrect tool calls before deployment unmet.

D:

Both steps target different results—reject tool arguments that fail a declared shape before tool execution and reduce repeated TLS setup overhead for high-volume application traffic—because a strict schema constrains the arguments a model may emit, but application authorization and business-rule checks remain necessary and connection reuse lowers transport overhead but still requires limits and health handling for long-lived clients; the required capabilities are instead to catch unsafe credentials and incorrect tool calls before deployment and to avoid retrying a malformed request that the API consistently rejects.

Question 773

A team maintaining a multimodal customer portal must meet 2 separate acceptance criteria: (1) ask Claude to interpret a supported image supplied as part of a request; (2) ensure refunds above a policy limit receive human sign-off. Which two implementation actions directly satisfy the stated requirements? Choose TWO.

  1. Record a blameless timeline, root causes, and preventive controls.
  2. Add application-owned approval logic around the refund tool.
  3. Release to a small monitored cohort before broad deployment.
  4. Validate generated fields against domain rules after parsing.
  5. Include a typed image content block in the message.

Explanation

Correct Answers

B:

The test to ensure refunds above a policy limit receive human sign-off is met because decisions that create financial side effects require deterministic authorization checks beyond a model's recommendation; the remaining selection covers the separate need to ask Claude to interpret a supported image supplied as part of a request.

E:

The test to ask Claude to interpret a supported image supplied as part of a request is met because multimodal input must use the API's image representation and supported sourcing; merely writing a filename in text does not transfer pixels; the remaining selection covers the separate need to ensure refunds above a policy limit receive human sign-off.

Incorrect Answers

A:

The action is useful for the separate capability to learn from a tool-side effect caused by an unexpected model action because post-incident review should identify inadequate approvals and monitoring and turn findings into testable changes, rather than blaming the language model; however, the required test outcomes are to ensure refunds above a policy limit receive human sign-off and to ask Claude to interpret a supported image supplied as part of a request.

C:

The action is useful for the separate capability to limit the blast radius of a new model configuration during rollout because canary exposure provides real traffic signals with an option to stop or roll back, unlike a simultaneous change for every customer; however, the required test outcomes are to ensure refunds above a policy limit receive human sign-off and to ask Claude to interpret a supported image supplied as part of a request.

D:

The action is useful for the separate capability to stop a well-formed JSON response from introducing an impossible customer status because a syntactically valid object may still contain invalid business values; server-side semantics determine whether it is acceptable; however, the required test outcomes are to ensure refunds above a policy limit receive human sign-off and to ask Claude to interpret a supported image supplied as part of a request.

Question 774

A team maintaining a Python Claude API backend has two different engineering requirements: (1) detect rising inference errors after a feature reaches production, and (2) catch mismatches between model calls, tool handlers, and application persistence. Which TWO independent changes directly meet them? Choose TWO.

  1. Route Claude requests through a trusted backend service.
  2. Exercise complete request-to-tool-to-response paths in a controlled environment.
  3. Track latency, failure rates, token use, and output-quality signals over time.
  4. Keep versioned artifacts and a tested rollback procedure.
  5. Use the API's documented POST request with a JSON body.

Explanation

Correct Answers

B:

To catch mismatches between model calls, tool handlers, and application persistence as part of a design that must also detect rising inference errors after a feature reaches production, unit tests cannot reveal every serialization or permission issue across service boundaries; integration tests validate the connected contract; this is why the action is one of the two required selections.

C:

To detect rising inference errors after a feature reaches production as part of a design that must also catch mismatches between model calls, tool handlers, and application persistence, production health spans infrastructure and model behavior, so a healthy HTTP server alone does not prove the AI feature is functioning; this is why the action is one of the two required selections.

Incorrect Answers

A:

This alternative serves the separate goal to keep a browser application from exposing the secret used for model calls because a backend controls credentials and access policy while the UI sends user input through an authenticated application endpoint; the required outcomes are instead to detect rising inference errors after a feature reaches production and to catch mismatches between model calls, tool handlers, and application persistence.

D:

This alternative serves the separate goal to restore a known-working prompt and model configuration after quality degrades because model IDs, prompts, schemas, and tool contracts are release components; restoring just the application binary may not restore prior behavior; the required outcomes are instead to detect rising inference errors after a feature reaches production and to catch mismatches between model calls, tool handlers, and application persistence.

E:

This alternative serves the separate goal to send a model inference request with the operation expected by the REST endpoint because a state-changing inference operation is submitted as an HTTP request with an explicit method and payload, not inferred from a browser URL alone; the required outcomes are instead to detect rising inference errors after a feature reaches production and to catch mismatches between model calls, tool handlers, and application persistence.

Question 775

In a customer-facing AI feature, testing reveals that the system cannot test experimental prompts and tools without reaching real customer records and cannot avoid an outage when a pinned model version approaches retirement. Select the two targeted actions that close these gaps. Choose TWO.

  1. Keep versioned artifacts and a tested rollback procedure.
  2. Inspect response.content blocks and metadata through the SDK's typed representation.
  3. Track provider deprecation notices and test migration before the removal date.
  4. Exercise complete request-to-tool-to-response paths in a controlled environment.
  5. Separate development credentials, data, and tool endpoints from production.

Explanation

Correct Answers

C:

To avoid an outage when a pinned model version approaches retirement as part of a design that must also test experimental prompts and tools without reaching real customer records, lifecycle management treats model availability as a dependency and validates output regressions before switching production traffic; this is why the action is one of the two required selections.

E:

To test experimental prompts and tools without reaching real customer records as part of a design that must also avoid an outage when a pinned model version approaches retirement, environment isolation prevents exploratory code from inheriting real business permissions and improves reproducibility of failures; this is why the action is one of the two required selections.

Incorrect Answers

A:

This alternative serves the separate goal to restore a known-working prompt and model configuration after quality degrades because model IDs, prompts, schemas, and tool contracts are release components; restoring just the application binary may not restore prior behavior; the required outcomes are instead to test experimental prompts and tools without reaching real customer records and to avoid an outage when a pinned model version approaches retirement.

B:

This alternative serves the separate goal to avoid assuming a structured SDK response is a plain string because the SDK exposes block types, stop reason, and usage separately; naïve string concatenation can lose tool requests or other content; the required outcomes are instead to test experimental prompts and tools without reaching real customer records and to avoid an outage when a pinned model version approaches retirement.

D:

This alternative serves the separate goal to catch mismatches between model calls, tool handlers, and application persistence because unit tests cannot reveal every serialization or permission issue across service boundaries; integration tests validate the connected contract; the required outcomes are instead to test experimental prompts and tools without reaching real customer records and to avoid an outage when a pinned model version approaches retirement.

Question 776

The owner of a production inference application needs to trust a callback only after verifying who actually sent it as well as make a direct REST request with a declared supported API contract. Which two changes belong in the minimum remediation plan? Choose TWO.

  1. Use asynchronous SDK calls with bounded concurrency.
  2. Supply the required API version header.
  3. Authenticate incoming callbacks and validate their signatures where supported.
  4. Retry eligible failures with capped exponential backoff and jitter.
  5. Version prompt files alongside relevant application code.

Explanation

Correct Answers

B:

To make a direct REST request with a declared supported API contract as part of a design that must also trust a callback only after verifying who actually sent it, the version header selects the protocol contract supported by the endpoint, preventing accidental dependence on unspecified behavior; this is why the action is one of the two required selections.

C:

To trust a callback only after verifying who actually sent it as part of a design that must also make a direct REST request with a declared supported API contract, network arrival alone does not prove a webhook's origin; authenticity checks prevent forged events from changing agent state; this is why the action is one of the two required selections.

Incorrect Answers

A:

This alternative serves the separate goal to serve many concurrent network-bound Claude requests without blocking worker threads unnecessarily because async I/O can overlap network waits, but it does not remove provider limits or make CPU-heavy postprocessing nonblocking; the required outcomes are instead to trust a callback only after verifying who actually sent it and to make a direct REST request with a declared supported API contract.

D:

This alternative serves the separate goal to recover transient 429 or 5xx responses without creating a retry storm because backoff spreads retry traffic and respects transient capacity limits; blind immediate repetition worsens overload; the required outcomes are instead to trust a callback only after verifying who actually sent it and to make a direct REST request with a declared supported API contract.

E:

This alternative serves the separate goal to trace exactly which prompt template was deployed with a code change because prompts change software behavior and should be reviewed and diffed like code, rather than edited anonymously in production; the required outcomes are instead to trust a callback only after verifying who actually sent it and to make a direct REST request with a declared supported API contract.

Question 777

Two approval criteria are still open for a business workflow that calls Claude: the team must turn a support desk's responsiveness expectation into an engineering objective and state how much interruption a customer-facing AI feature can tolerate. Which two implementation actions satisfy them? Choose TWO.

  1. Define availability objectives with failure and recovery criteria.
  2. Detect transport interruption and mark incomplete generated content.
  3. Set an end-to-end latency target and a measurement method.
  4. List upstream systems with permissions, quotas, and failure behavior.
  5. Specify expected concurrency and enforce admission or queueing limits.

Explanation

Correct Answers

A:

To state how much interruption a customer-facing AI feature can tolerate as part of a design that must also turn a support desk's responsiveness expectation into an engineering objective, reliability requirements determine fallback, retries, health monitoring, and degraded behavior; they are distinct from the response's linguistic quality; this is why the action is one of the two required selections.

C:

To turn a support desk's responsiveness expectation into an engineering objective as part of a design that must also state how much interruption a customer-facing AI feature can tolerate, a latency target is a nonfunctional requirement; model choice, network overhead, tool calls, and rendering all affect whether the service meets it; this is why the action is one of the two required selections.

Incorrect Answers

B:

This alternative serves the separate goal to avoid silently treating a broken streaming connection as a complete answer because network termination before a final message event can leave a partial response, which needs explicit recovery rather than success handling; the required outcomes are instead to turn a support desk's responsiveness expectation into an engineering objective and to state how much interruption a customer-facing AI feature can tolerate.

D:

This alternative serves the separate goal to identify which search services and internal systems are critical to the workflow because a dependency inventory exposes the operational and authorization constraints that a purely model-centered design misses; the required outcomes are instead to turn a support desk's responsiveness expectation into an engineering objective and to state how much interruption a customer-facing AI feature can tolerate.

E:

This alternative serves the separate goal to design for a documented burst of simultaneous requests without uncontrolled overload because capacity planning separates average traffic from peak demand and prevents upstream rate limits from becoming cascading application failures; the required outcomes are instead to turn a support desk's responsiveness expectation into an engineering objective and to state how much interruption a customer-facing AI feature can tolerate.

Question 778

For a high-volume Messages API gateway, the missing capability is to design for a documented burst of simultaneous requests without uncontrolled overload. What should the developers implement?

  1. Place stable application rules in system instructions and untrusted requests in user content.
  2. Convert quality expectations into labeled tests and failure thresholds.
  3. Specify expected concurrency and enforce admission or queueing limits.
  4. Define measurable answer-format and citation acceptance criteria.

Explanation

Correct Answer

C:

To design for a documented burst of simultaneous requests without uncontrolled overload in this high-volume Messages API gateway, capacity planning separates average traffic from peak demand and prevents upstream rate limits from becoming cascading application failures; the proposed implementation directly supplies that capability.

Incorrect Answers

A:

The proposed alternative helps to keep application policy separate from a user's question because instruction placement preserves the distinction between developer intent and user-supplied data; concatenating everything blurs that boundary; that mechanism does not meet the different requirement to design for a documented burst of simultaneous requests without uncontrolled overload.

B:

The proposed alternative helps to verify that generated categorization meets a business-approved standard because acceptance is meaningful only with representative cases and explicit quality criteria; subjective demo approval cannot establish production fitness; that mechanism does not meet the different requirement to design for a documented burst of simultaneous requests without uncontrolled overload.

D:

The proposed alternative helps to translate a request for citation-backed answers into observable application behavior because a functional requirement specifies what the user must receive and how success is evaluated, not merely that the team will use a language model; that mechanism does not meet the different requirement to design for a documented burst of simultaneous requests without uncontrolled overload.

Question 779

An operational review of a hosted document-processing service reports distinct failures: it cannot send a model inference request with the operation expected by the REST endpoint, and it cannot detect a changed tool response field before it breaks downstream code. Which two repairs are required? Choose TWO.

  1. Refactor in bounded steps under existing tests.
  2. Add automated consumer-provider contract tests.
  3. Place a valid secret in the documented API authentication header.
  4. Exercise complete request-to-tool-to-response paths in a controlled environment.
  5. Use the API's documented POST request with a JSON body.

Explanation

Correct Answers

B:

To detect a changed tool response field before it breaks downstream code as part of a design that must also send a model inference request with the operation expected by the REST endpoint, schema contracts make integration changes visible and prevent a renamed field from silently corrupting later model context; this is why the action is one of the two required selections.

E:

To send a model inference request with the operation expected by the REST endpoint as part of a design that must also detect a changed tool response field before it breaks downstream code, a state-changing inference operation is submitted as an HTTP request with an explicit method and payload, not inferred from a browser URL alone; this is why the action is one of the two required selections.

Incorrect Answers

A:

This alternative serves the separate goal to preserve behavior while replacing a fragile parser in an AI integration because small changes constrain regression scope; a wholesale rewrite makes it harder to isolate whether model or parsing behavior changed; the required outcomes are instead to send a model inference request with the operation expected by the REST endpoint and to detect a changed tool response field before it breaks downstream code.

C:

This alternative serves the separate goal to authenticate a server-to-server Claude API request because API authentication is a request transport concern; placing a key inside the prompt body neither authenticates the request nor protects it; the required outcomes are instead to send a model inference request with the operation expected by the REST endpoint and to detect a changed tool response field before it breaks downstream code.

D:

This alternative serves the separate goal to catch mismatches between model calls, tool handlers, and application persistence because unit tests cannot reveal every serialization or permission issue across service boundaries; integration tests validate the connected contract; the required outcomes are instead to send a model inference request with the operation expected by the REST endpoint and to detect a changed tool response field before it breaks downstream code.

Question 780

An architecture decision for a TypeScript Claude integration must support distinguish API transport failures from a successfully returned model refusal and detect that the assistant has naturally finished a response. Which option gives a complete rather than partial solution?

  1. Perform both steps: (1) Inspect HTTP status and structured error payload before interpreting model output; (2) detect refusal and apply the product's allowed fallback or messaging policy.
  2. Perform both steps: (1) Inspect HTTP status and structured error payload before interpreting model output; (2) treat stop_reason end_turn as a completed assistant turn.
  3. Perform both steps: (1) Include the necessary conversation history in each Messages API call; (2) treat stop_reason end_turn as a completed assistant turn.
  4. Perform both steps: (1) Include the necessary conversation history in each Messages API call; (2) detect refusal and apply the product's allowed fallback or messaging policy.

Explanation

Correct Answer

B:

The first change enables the team to distinguish API transport failures from a successfully returned model refusal because rate limits, authorization failures, and server errors happen at the request layer, unlike a normal response with a refusal stop reason; the second addresses the separate need to detect that the assistant has naturally finished a response because the completion signal is explicit in the response metadata; it should not be inferred solely from a sentence ending in punctuation.

Incorrect Answers

A:

The first step is appropriate for the requirement to distinguish API transport failures from a successfully returned model refusal because rate limits, authorization failures, and server errors happen at the request layer, unlike a normal response with a refusal stop reason; the other step instead aims to handle a safety refusal as a distinct application outcome because a refusal is not the same as a network failure; blindly retrying identical input may not produce an appropriate user experience, which does not meet the second requirement to detect that the assistant has naturally finished a response.

C:

The second step meets the requirement to detect that the assistant has naturally finished a response because the completion signal is explicit in the response metadata; it should not be inferred solely from a sentence ending in punctuation; the first step would only help to continue a chat with relevant prior turns despite the API being stateless because a response does not create server-side conversational memory by default; the client supplies retained turns and controls truncation, leaving the separate requirement to distinguish API transport failures from a successfully returned model refusal unmet.

D:

Both steps target different results—continue a chat with relevant prior turns despite the API being stateless and handle a safety refusal as a distinct application outcome—because a response does not create server-side conversational memory by default; the client supplies retained turns and controls truncation and a refusal is not the same as a network failure; blindly retrying identical input may not produce an appropriate user experience; the required capabilities are instead to distinguish API transport failures from a successfully returned model refusal and to detect that the assistant has naturally finished a response.

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