Question 450
A change request for a high-volume Messages API gateway combines two needs: distinguish queued, running, waiting-for-approval, failed, and completed agent jobs, and reject tool arguments that fail a declared shape before tool execution. Which implementation pair addresses both, not just one?
- Perform both steps: (1) Persist explicit job states with allowed transitions; (2) version tool input schemas and coordinate dependent client changes.
- Perform both steps: (1) Queue or reject excess requests using bounded admission control; (2) use strict tool input validation for the exposed tool interface.
- Perform both steps: (1) Queue or reject excess requests using bounded admission control; (2) version tool input schemas and coordinate dependent client changes.
- Perform both steps: (1) Persist explicit job states with allowed transitions; (2) use strict tool input validation for the exposed tool interface.
Explanation
Correct Answer
D:The first change enables the team 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; the second addresses the separate need 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.
Incorrect Answers
A:The first step is appropriate for the requirement 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; the other step instead aims to migrate tool argument fields without breaking older workers because contract versioning prevents producers and consumers from silently disagreeing about required fields or meanings, which does not meet the second requirement to reject tool arguments that fail a declared shape before tool execution.
B:The second step meets the requirement 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; the first step would only help to shield customer traffic when inference capacity is exhausted because controlled queuing protects system latency and prevents unbounded retries from overwhelming downstream provider quotas, leaving the separate requirement to distinguish queued, running, waiting-for-approval, failed, and completed agent jobs unmet.
C:Both steps target different results—shield customer traffic when inference capacity is exhausted and migrate tool argument fields without breaking older workers—because controlled queuing protects system latency and prevents unbounded retries from overwhelming downstream provider quotas and contract versioning prevents producers and consumers from silently disagreeing about required fields or meanings; the required capabilities are instead to distinguish queued, running, waiting-for-approval, failed, and completed agent jobs and to reject tool arguments that fail a declared shape before tool execution.
Question 451
The team responsible for a latency-sensitive model gateway needs to protect model API traffic against an impersonated network endpoint. What is the most direct engineering change?
- Compare both candidates on identical representative eval cases.
- Keep HTTPS certificate and hostname verification enabled.
- Route simple requests to a cheaper evaluated tier and escalate difficult ones.
- Iterate the SDK or HTTP event stream until its terminal event.
Explanation
Correct Answer
B:To protect model API traffic against an impersonated network endpoint in this latency-sensitive model gateway, verified TLS authenticates the service and protects data in transit; disabling validation defeats a key transport security property; the proposed implementation directly supplies that capability.
Incorrect Answers
A:The proposed alternative helps to choose between two supported model releases using evidence rather than reputation because holding inputs and scoring constant reveals quality and latency changes that generic model descriptions cannot settle; that mechanism does not meet the different requirement to protect model API traffic against an impersonated network endpoint.
C:The proposed alternative helps to minimize average model cost when most requests are easy but some require deeper reasoning because cascading matches compute to task difficulty but requires a reliable escalation rule and measurement of misrouting; that mechanism does not meet the different requirement to protect model API traffic against an impersonated network endpoint.
D:The proposed alternative helps to process partial model output as it is delivered over a streaming connection because streaming clients must handle both content deltas and final status rather than assuming the first chunk is the complete response; that mechanism does not meet the different requirement to protect model API traffic against an impersonated network endpoint.
Question 452
A test for a specialist-agent orchestration service fails because it cannot execute a requested internal business function inside the application's trusted process. Which change fixes the identified failure?
- Add a deterministic pre-execution hook for the sensitive operation.
- Inspect the tool_use block, invoke the authorized handler, and return tool_result.
- Consume the agent runtime's streamed event sequence.
- Define and validate a typed schema at the handoff.
Explanation
Correct Answer
B:To execute a requested internal business function inside the application's trusted process in this specialist-agent orchestration service, client tools are executed by the application, not by merely declaring a schema; every returned result must be matched to its tool-use identifier; the proposed implementation directly supplies that capability.
Incorrect Answers
A:The proposed alternative helps to block an unsafe shell or file operation before a tool executes because a pre-tool boundary can reject a forbidden action independently of model judgment; enforcement must occur before side effects happen; that mechanism does not meet the different requirement to execute a requested internal business function inside the application's trusted process.
C:The proposed alternative helps to deliver intermediate agent events to the client as they occur because event-driven consumption lets the application display progress and inspect tool actions without waiting for the entire multi-step run to finish; that mechanism does not meet the different requirement to execute a requested internal business function inside the application's trusted process.
D:The proposed alternative helps to pass reliable machine-readable outputs between independently developed agent stages because typed outputs make downstream contracts inspectable and reduce ambiguity caused by free-form prose passed as control data; that mechanism does not meet the different requirement to execute a requested internal business function inside the application's trusted process.
Question 453
A design review for an enterprise agent deployment asks for support to prevent an adversarial user from bypassing safety policy by roleplay instructions, together with a way to detect unauthorized modifications to a tool result before it reaches the model. Select the two appropriate changes. Choose TWO.
- Pause execution for explicit approval at the high-impact action boundary.
- Verify origin and integrity of sensitive external data where supported.
- Treat retrieved content as untrusted and prevent its instructions from controlling tools.
- Apply an input-side policy gate and route flagged cases to a safe path.
- Use layered policy enforcement and restrict powerful tools independently of chat text.
Explanation
Correct Answers
B:To detect unauthorized modifications to a tool result before it reaches the model as part of a design that must also prevent an adversarial user from bypassing safety policy by roleplay instructions, trusted decisions require confidence that tool outputs were not altered; transport and signature checks complement content validation; this is why the action is one of the two required selections.
E:To prevent an adversarial user from bypassing safety policy by roleplay instructions as part of a design that must also detect unauthorized modifications to a tool result before it reaches the model, adversarial prompting can target instruction hierarchy, so product controls and tool authorization must remain effective when wording is manipulated; this is why the action is one of the two required selections.
Incorrect Answers
A:This alternative serves the separate goal to require a person to approve an irreversible administrative change because human approval is a policy-enforced transition, not a note written in the model's plan or a message displayed afterward; the required outcomes are instead to prevent an adversarial user from bypassing safety policy by roleplay instructions and to detect unauthorized modifications to a tool result before it reaches the model.
C:This alternative serves the separate goal to resist instructions embedded in a retrieved support ticket that demand secret disclosure because source material is data rather than authority; enforcing tool permissions and separating roles limits an attacker's ability to redirect execution; the required outcomes are instead to prevent an adversarial user from bypassing safety policy by roleplay instructions and to detect unauthorized modifications to a tool result before it reaches the model.
D:This alternative serves the separate goal to identify prohibited requests before an automated agent invokes business tools because early filtering reduces inappropriate work, but later checks remain necessary because tool outputs can also introduce risks; the required outcomes are instead to prevent an adversarial user from bypassing safety policy by roleplay instructions and to detect unauthorized modifications to a tool result before it reaches the model.
Question 454
For a Python Claude API backend, the missing capability is to preserve an understandable outcome when model generation cannot complete. What should the developers implement?
- Use the API's documented POST request with a JSON body.
- Exercise complete request-to-tool-to-response paths in a controlled environment.
- Retry eligible failures with capped exponential backoff and jitter.
- Specify user-visible degraded responses and recovery paths.
Explanation
Correct Answer
D:To preserve an understandable outcome when model generation cannot complete in this Python Claude API backend, a fallback requirement defines what customers see when an API call times out or is refused rather than assuming every inference will succeed; the proposed implementation directly supplies that capability.
Incorrect Answers
A:The proposed alternative helps 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; that mechanism does not meet the different requirement to preserve an understandable outcome when model generation cannot complete.
B:The proposed alternative helps 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; that mechanism does not meet the different requirement to preserve an understandable outcome when model generation cannot complete.
C:The proposed alternative helps 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; that mechanism does not meet the different requirement to preserve an understandable outcome when model generation cannot complete.
Question 455
A production readiness review of a long-context summarization workflow finds that the team cannot yet provide a document's necessary meaning without disclosing irrelevant personal identifiers or stop verbose tool payloads from consuming most of a long-running agent's context. Which paired changes directly remedy the two gaps?
- Perform both steps: (1) Minimize or redact sensitive fields before constructing context; (2) retain essential fields and discard irrelevant tool output from active context.
- Perform both steps: (1) Escape or sanitize model text in the output presentation layer; (2) map or reject values not in the approved enumeration.
- Perform both steps: (1) Escape or sanitize model text in the output presentation layer; (2) retain essential fields and discard irrelevant tool output from active context.
- Perform both steps: (1) Minimize or redact sensitive fields before constructing context; (2) map or reject values not in the approved enumeration.
Explanation
Correct Answer
A:The first change enables the team to provide a document's necessary meaning without disclosing irrelevant personal identifiers because data minimization reduces the sensitive information transmitted for inference while retaining the facts the task legitimately requires; the second addresses the separate need to stop verbose tool payloads from consuming most of a long-running agent's context because summarizing or filtering observations keeps the reasoning window focused without deleting required facts from durable application state.
Incorrect Answers
B:Both steps target different results—safely render generated text in a browser without executing embedded markup and prevent a generated status outside a database's accepted code set—because generated content is untrusted for HTML rendering; injection into the DOM must not become executable client-side code and controlled vocabularies need explicit allowed values; a plausible new status label can still break downstream contracts; the required capabilities are instead to provide a document's necessary meaning without disclosing irrelevant personal identifiers and to stop verbose tool payloads from consuming most of a long-running agent's context.
C:The second step meets the requirement to stop verbose tool payloads from consuming most of a long-running agent's context because summarizing or filtering observations keeps the reasoning window focused without deleting required facts from durable application state; the first step would only help to safely render generated text in a browser without executing embedded markup because generated content is untrusted for HTML rendering; injection into the DOM must not become executable client-side code, leaving the separate requirement to provide a document's necessary meaning without disclosing irrelevant personal identifiers unmet.
D:The first step is appropriate for the requirement to provide a document's necessary meaning without disclosing irrelevant personal identifiers because data minimization reduces the sensitive information transmitted for inference while retaining the facts the task legitimately requires; the other step instead aims to prevent a generated status outside a database's accepted code set because controlled vocabularies need explicit allowed values; a plausible new status label can still break downstream contracts, which does not meet the second requirement to stop verbose tool payloads from consuming most of a long-running agent's context.
Question 456
An end-to-end test for an agent access-control assessment checks whether it can stop a coding agent from running commands that match a prohibited destructive pattern and avoid accidentally allowing critical writes when a safety check crashes. Which plan should be approved to satisfy both checks?
- Perform both steps: (1) Enforce a pre-tool hook that denies the disallowed call; (2) define fail-closed behavior for an enforcement hook and test failure paths.
- Perform both steps: (1) Enforce a pre-tool hook that denies the disallowed call; (2) grant each runtime tool only the operations and resources required for its task.
- Perform both steps: (1) Record result metadata in a post-tool hook; (2) grant each runtime tool only the operations and resources required for its task.
- Perform both steps: (1) Record result metadata in a post-tool hook; (2) define fail-closed behavior for an enforcement hook and test failure paths.
Explanation
Correct Answer
A:The first change enables the team to stop a coding agent from running commands that match a prohibited destructive pattern because a deterministic hook checks the proposed command before execution, providing a stronger control than asking the agent to self-police; the second addresses the separate need to avoid accidentally allowing critical writes when a safety check crashes because a guardrail that silently skips checks on errors is not an effective boundary; failures must prevent the protected action when policy requires it.
Incorrect Answers
B:The first step is appropriate for the requirement to stop a coding agent from running commands that match a prohibited destructive pattern because a deterministic hook checks the proposed command before execution, providing a stronger control than asking the agent to self-police; the other step instead aims to limit the consequences of a mistaken tool call in a production support system because least-privilege permissions shrink the blast radius of model errors and compromised prompts, especially for mutating actions, which does not meet the second requirement to avoid accidentally allowing critical writes when a safety check crashes.
C:Both steps target different results—capture outcomes of agent command execution for security review and limit the consequences of a mistaken tool call in a production support system—because post-execution hooks can log and inspect what happened; they are unsuitable as the sole barrier to a side effect that already occurred and least-privilege permissions shrink the blast radius of model errors and compromised prompts, especially for mutating actions; the required capabilities are instead to stop a coding agent from running commands that match a prohibited destructive pattern and to avoid accidentally allowing critical writes when a safety check crashes.
D:The second step meets the requirement to avoid accidentally allowing critical writes when a safety check crashes because a guardrail that silently skips checks on errors is not an effective boundary; failures must prevent the protected action when policy requires it; the first step would only help to capture outcomes of agent command execution for security review because post-execution hooks can log and inspect what happened; they are unsuitable as the sole barrier to a side effect that already occurred, leaving the separate requirement to stop a coding agent from running commands that match a prohibited destructive pattern unmet.
Question 457
For a shared tool gateway for Claude, a design is acceptable only if it can make a repeatable multi-step code-review methodology available to an agent while also being able to invoke a proprietary internal pricing database unavailable through a built-in feature. Which paired implementation is sufficient?
- Perform both steps: (1) Create a Skill containing the procedure and supporting instructions; (2) write a custom tool whose handler runs in the application.
- Perform both steps: (1) Use the MCP stdio transport with properly framed protocol messages; (2) write a custom tool whose handler runs in the application.
- Perform both steps: (1) Create a Skill containing the procedure and supporting instructions; (2) publish a properly described mcp tool and its input contract.
- Perform both steps: (1) Use the MCP stdio transport with properly framed protocol messages; (2) publish a properly described mcp tool and its input contract.
Explanation
Correct Answer
A:The first change enables the team to make a repeatable multi-step code-review methodology available to an agent because a Skill teaches how to perform a task, whereas a tool is the executable capability that performs an external operation; the second addresses the separate need to invoke a proprietary internal pricing database unavailable through a built-in feature because custom tools expose application-specific actions while retaining control over authentication, validation, and network access.
Incorrect Answers
B:The second step meets the requirement to invoke a proprietary internal pricing database unavailable through a built-in feature because custom tools expose application-specific actions while retaining control over authentication, validation, and network access; the first step would only help to launch a local MCP server as a subprocess managed by a compatible client because standard input and output suit local process integration, where protocol output must not be mixed with arbitrary logging noise, leaving the separate requirement to make a repeatable multi-step code-review methodology available to an agent unmet.
C:The first step is appropriate for the requirement to make a repeatable multi-step code-review methodology available to an agent because a Skill teaches how to perform a task, whereas a tool is the executable capability that performs an external operation; the other step instead aims to let an MCP client invoke a business capability through a standard server interface because MCP tools represent callable operations with typed parameters; the server executes authorized behavior rather than only providing reference text, which does not meet the second requirement to invoke a proprietary internal pricing database unavailable through a built-in feature.
D:Both steps target different results—launch a local MCP server as a subprocess managed by a compatible client and let an MCP client invoke a business capability through a standard server interface—because standard input and output suit local process integration, where protocol output must not be mixed with arbitrary logging noise and MCP tools represent callable operations with typed parameters; the server executes authorized behavior rather than only providing reference text; the required capabilities are instead to make a repeatable multi-step code-review methodology available to an agent and to invoke a proprietary internal pricing database unavailable through a built-in feature.
Question 458
A planned upgrade to a token-consumption optimization sprint has two requirements that must both pass: choose between two supported model releases using evidence rather than reputation; preserve critical service during a temporary primary-model problem. Identify the two direct implementations. Choose TWO.
- Revalidate prompts, schemas, safety constraints, and tools after model migration.
- Design an approved fallback model path with compatibility checks.
- Evaluate a Sonnet-family model against a realistic task set.
- Model the response as conditioned token generation.
- Compare both candidates on identical representative eval cases.
Explanation
Correct Answers
B:To preserve critical service during a temporary primary-model problem as part of a design that must also choose between two supported model releases using evidence rather than reputation, fallbacks improve resilience only when downstream schemas, cost limits, and policy controls work with the alternate model; this is why the action is one of the two required selections.
E:To choose between two supported model releases using evidence rather than reputation as part of a design that must also preserve critical service during a temporary primary-model problem, holding inputs and scoring constant reveals quality and latency changes that generic model descriptions cannot settle; this is why the action is one of the two required selections.
Incorrect Answers
A:This alternative serves the separate goal to avoid assuming a model-version change preserves tool output and writing behavior because new releases can change response distributions or supported behaviors, making a model upgrade a real software change; the required outcomes are instead to choose between two supported model releases using evidence rather than reputation and to preserve critical service during a temporary primary-model problem.
C:This alternative serves the separate goal to deliver moderately complex development assistance with balanced quality, cost, and latency because a balanced model tier often offers useful capability without the latency or cost of the most expensive tier; evaluation should drive final selection; the required outcomes are instead to choose between two supported model releases using evidence rather than reputation and to preserve critical service during a temporary primary-model problem.
D:This alternative serves the separate goal to explain why language-model output is generated step by step rather than fetched from a fixed response table because autoregressive decoding selects subsequent tokens based on supplied context and prior output, which creates flexible but non-guaranteed wording; the required outcomes are instead to choose between two supported model releases using evidence rather than reputation and to preserve critical service during a temporary primary-model problem.
Question 459
An operational review of a hosted document-processing service reports distinct failures: it cannot recover transient 429 or 5xx responses without creating a retry storm, and it cannot ensure a rebuilt agent uses the same tested SDK dependency set. Which two repairs are required? Choose TWO.
- Treat model migration as a tested deployment with staged traffic.
- Refactor in bounded steps under existing tests.
- Commit lockfiles and control library upgrades.
- Retry eligible failures with capped exponential backoff and jitter.
- Require peer review focused on API, security, and tool-side effects.
Explanation
Correct Answers
C:To ensure a rebuilt agent uses the same tested SDK dependency set as part of a design that must also recover transient 429 or 5xx responses without creating a retry storm, dependency pinning reduces environment drift; transitive SDK updates can change behavior even when prompt text is unchanged; this is why the action is one of the two required selections.
D:To recover transient 429 or 5xx responses without creating a retry storm as part of a design that must also ensure a rebuilt agent uses the same tested SDK dependency set, backoff spreads retry traffic and respects transient capacity limits; blind immediate repetition worsens overload; this is why the action is one of the two required selections.
Incorrect Answers
A:This alternative serves the separate goal to compare candidate models before switching a mission-critical endpoint because model upgrades may alter output distributions or supported features; staged validation reduces the chance of an unnoticed behavior shift; the required outcomes are instead to recover transient 429 or 5xx responses without creating a retry storm and to ensure a rebuilt agent uses the same tested SDK dependency set.
B: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 recover transient 429 or 5xx responses without creating a retry storm and to ensure a rebuilt agent uses the same tested SDK dependency set.
E:This alternative serves the separate goal 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 required outcomes are instead to recover transient 429 or 5xx responses without creating a retry storm and to ensure a rebuilt agent uses the same tested SDK dependency set.
Question 460
Two approval criteria are still open for a secure tool-execution boundary: the team must secure an agent against both malicious retrieved text and unsafe tool execution and require a person to approve an irreversible administrative change. Which two implementation actions satisfy them? Choose TWO.
- Apply an input-side policy gate and route flagged cases to a safe path.
- Validate or moderate final content before release.
- Pause execution for explicit approval at the high-impact action boundary.
- Combine prompt boundaries, scoped credentials, validation, and approval gates.
- Use layered policy enforcement and restrict powerful tools independently of chat text.
Explanation
Correct Answers
C:To require a person to approve an irreversible administrative change as part of a design that must also secure an agent against both malicious retrieved text and unsafe tool execution, human approval is a policy-enforced transition, not a note written in the model's plan or a message displayed afterward; this is why the action is one of the two required selections.
D:To secure an agent against both malicious retrieved text and unsafe tool execution as part of a design that must also require a person to approve an irreversible administrative change, defense in depth assumes individual filters can fail; independent enforcement boundaries reduce the chance of harmful side effects; this is why the action is one of the two required selections.
Incorrect Answers
A:This alternative serves the separate goal to identify prohibited requests before an automated agent invokes business tools because early filtering reduces inappropriate work, but later checks remain necessary because tool outputs can also introduce risks; the required outcomes are instead to secure an agent against both malicious retrieved text and unsafe tool execution and to require a person to approve an irreversible administrative change.
B:This alternative serves the separate goal to prevent generated advice that violates the product's publication rules from being shown because output checks catch issues emerging during generation; they complement rather than replace input controls and tool restrictions; the required outcomes are instead to secure an agent against both malicious retrieved text and unsafe tool execution and to require a person to approve an irreversible administrative change.
E:This alternative serves the separate goal to prevent an adversarial user from bypassing safety policy by roleplay instructions because adversarial prompting can target instruction hierarchy, so product controls and tool authorization must remain effective when wording is manipulated; the required outcomes are instead to secure an agent against both malicious retrieved text and unsafe tool execution and to require a person to approve an irreversible administrative change.