Question 660
For a subscription support service, a reviewer identifies two independent obligations: a changed account record supersedes an old conversational value and an ordinary in-policy question is not escalated needlessly. Which proposal should be approved?
- reuse the value from the first turn without checking for change; and reserve human handoff for specified triggers and let normal cases proceed
- refresh the authoritative source before a consequential write; and reserve human handoff for specified triggers and let normal cases proceed
- reuse the value from the first turn without checking for change; and send every low-risk question to a human as a default
- refresh the authoritative source before a consequential write; and send every low-risk question to a human as a default
Explanation
Correct Answer
B:In a subscription support service, current external state protects against stale facts preserved in context; alongside this, well-defined boundaries maintain autonomy without bypassing risky decisions, satisfying the distinct requirements that a changed account record supersedes an old conversational value and that an ordinary in-policy question is not escalated needlessly.
Incorrect Answers
A:In a subscription support service, well-defined boundaries maintain autonomy without bypassing risky decisions, but conversations do not automatically reflect later database updates; therefore the separate requirement that a changed account record supersedes an old conversational value would remain unsatisfied.
C:In a subscription support service, conversations do not automatically reflect later database updates and blanket escalation defeats the purpose of routine assistance; the design would fail both conditions: a changed account record supersedes an old conversational value and an ordinary in-policy question is not escalated needlessly.
D:In a subscription support service, current external state protects against stale facts preserved in context, but blanket escalation defeats the purpose of routine assistance; therefore the separate requirement that an ordinary in-policy question is not escalated needlessly would remain unsatisfied.
Question 661
Production readiness for a nightly code quality job depends on ensuring that a pull-request bot runs with minimum necessary authority and that regressions introduced during refinement are detected and fixed. Select the most appropriate combined approach.
- scope runner credentials, worktree access, and allowed tools to its task; and restore or revise the problematic change and retest affected paths
- implement only scope runner credentials, worktree access, and allowed tools to its task; make no corresponding change for the second requirement
- give the bot administrator credentials to avoid permission failures; and keep layering patches without checking the last stable behavior
- scope runner credentials, worktree access, and allowed tools to its task; and keep layering patches without checking the last stable behavior
- give the bot administrator credentials to avoid permission failures; and restore or revise the problematic change and retest affected paths
Explanation
Correct Answer
A:In a nightly code quality job, CI identity and tool permissions control real write capabilities independent of model instructions; alongside this, iteration includes detecting regressions and returning to a passing state, satisfying the distinct requirements that a pull-request bot runs with minimum necessary authority and that regressions introduced during refinement are detected and fixed.
Incorrect Answers
B:In a nightly code quality job, CI identity and tool permissions control real write capabilities independent of model instructions, but the independent mechanism for the other criterion is absent: iteration includes detecting regressions and returning to a passing state; as written, the design does not ensure that regressions introduced during refinement are detected and fixed.
C:In a nightly code quality job, broad credentials increase the blast radius of a compromised or mistaken job and cumulative changes can bury the source of failure; the design would fail both conditions: a pull-request bot runs with minimum necessary authority and regressions introduced during refinement are detected and fixed.
D:In a nightly code quality job, CI identity and tool permissions control real write capabilities independent of model instructions, but cumulative changes can bury the source of failure; therefore the separate requirement that regressions introduced during refinement are detected and fixed would remain unsatisfied.
E:In a nightly code quality job, iteration includes detecting regressions and returning to a passing state, but broad credentials increase the blast radius of a compromised or mistaken job; therefore the separate requirement that a pull-request bot runs with minimum necessary authority would remain unsatisfied.
Question 662
The architect for a structured case-file ingestion workflow needs to ensure that the assistant can use external service capabilities via MCP. What should the team implement?
- commit a reviewed project .mcp.json without credentials
- use a user-local MCP configuration rather than committing project configuration
- describe the external API in a prompt without connecting any tool
- configure a compatible MCP server and expose its declared tools to the client
Explanation
Correct Answer
D:For a structured case-file ingestion workflow, MCP bridges a client to provider-defined capabilities rather than inventing a Claude built-in; this directly establishes that the assistant can use external service capabilities via MCP instead of relying on an unrelated workflow control.
Incorrect Answers
A:Although project-scoped configuration can be distributed with the repository while keeping secrets external, that only supports the distinct goal that a team shares a repository-specific MCP server configuration, not the requirement that the assistant can use external service capabilities via MCP in a structured case-file ingestion workflow.
B:Although user scope is appropriate for personal endpoints and avoids distributing local paths, that only supports the distinct goal that one developer keeps an MCP server private to their machine, not the requirement that the assistant can use external service capabilities via MCP in a structured case-file ingestion workflow.
C:In a structured case-file ingestion workflow, instructions cannot create an executable integration endpoint; as a result, the stated requirement that the assistant can use external service capabilities via MCP would still not be met.
Question 663
Before expanding a literature synthesis workflow, the owner must demonstrate that partial data is not presented as a complete successful result, while also demonstrating that every high-impact tool invocation produces an audit event. Which design provides both controls?
- return an explicit partial status with missing-item identifiers; and record the tool name, permitted inputs, outcome, and task identifier in a hook
- return an explicit partial status with missing-item identifiers; and infer tool activity from the final natural-language answer
- silently omit records that failed to load; and infer tool activity from the final natural-language answer
- silently omit records that failed to load; and record the tool name, permitted inputs, outcome, and task identifier in a hook
Explanation
Correct Answer
A:In a literature synthesis workflow, failure metadata preserves the difference between a sparse valid response and incomplete execution; alongside this, runtime instrumentation observes actual proposed or completed tool activity, satisfying the distinct requirements that partial data is not presented as a complete successful result and that every high-impact tool invocation produces an audit event.
Incorrect Answers
B:In a literature synthesis workflow, failure metadata preserves the difference between a sparse valid response and incomplete execution, but a summary can omit failed calls or delegated actions; therefore the separate requirement that every high-impact tool invocation produces an audit event would remain unsatisfied.
C:In a literature synthesis workflow, the coordinator could publish an inaccurate completeness claim and a summary can omit failed calls or delegated actions; the design would fail both conditions: partial data is not presented as a complete successful result and every high-impact tool invocation produces an audit event.
D:In a literature synthesis workflow, runtime instrumentation observes actual proposed or completed tool activity, but the coordinator could publish an inaccurate completeness claim; therefore the separate requirement that partial data is not presented as a complete successful result would remain unsatisfied.
Question 664
During review of a pull-request review pipeline, an unmet criterion states that repeated delivery of one action request cannot create duplicate writes. Which measure resolves this specific gap?
- track individual stage outcomes and use explicit recovery or compensating actions
- blindly resubmit the same write after every timeout
- use an idempotency key and confirm the prior commit before retrying
- send a structured state packet with completed checks and unresolved decisions
Explanation
Correct Answer
C:For a pull-request review pipeline, side-effecting operations need deduplication against timeouts and repeated messages; this directly establishes that repeated delivery of one action request cannot create duplicate writes instead of relying on an unrelated workflow control.
Incorrect Answers
A:Although an orchestrated workflow must not report overall success while a required step failed, that only supports the distinct goal that partial completion of a multi-step operation is handled safely, not the requirement that repeated delivery of one action request cannot create duplicate writes in a pull-request review pipeline.
B:In a pull-request review pipeline, the first attempt might have committed despite a missing acknowledgment; as a result, the stated requirement that repeated delivery of one action request cannot create duplicate writes would still not be met.
D:Although a typed transition preserves obligations and facts between workflow stages, that only supports the distinct goal that a specialist handoff retains its prerequisites and evidence, not the requirement that repeated delivery of one action request cannot create duplicate writes in a pull-request review pipeline.
Question 665
In an engineering knowledge assistant, the acceptance criteria require that an existing file receives a precise localized replacement and that two teams' analyses are reconciled before publication. Which paired implementation meets both?
- overwrite the whole file with a newly generated copy; and publish whichever branch finishes last
- use Edit for an exact targeted change; and publish whichever branch finishes last
- implement only use Edit for an exact targeted change; make no corresponding change for the second requirement
- overwrite the whole file with a newly generated copy; and add a synthesis stage with explicit conflict-resolution criteria
- use Edit for an exact targeted change; and add a synthesis stage with explicit conflict-resolution criteria
- implement only add a synthesis stage with explicit conflict-resolution criteria; make no corresponding change for the first requirement
Explanation
Correct Answer
E:In an engineering knowledge assistant, surgical edits preserve unrelated lines and provide reviewable diffs; alongside this, merging divergent work requires an authoritative step rather than choosing by arrival time, satisfying the distinct requirements that an existing file receives a precise localized replacement and that two teams' analyses are reconciled before publication.
Incorrect Answers
A:In an engineering knowledge assistant, full replacement risks losing unrelated code and formatting and run order is not an evidence or quality standard; the design would fail both conditions: an existing file receives a precise localized replacement and two teams' analyses are reconciled before publication.
B:In an engineering knowledge assistant, surgical edits preserve unrelated lines and provide reviewable diffs, but run order is not an evidence or quality standard; therefore the separate requirement that two teams' analyses are reconciled before publication would remain unsatisfied.
C:In an engineering knowledge assistant, surgical edits preserve unrelated lines and provide reviewable diffs, but the independent mechanism for the other criterion is absent: merging divergent work requires an authoritative step rather than choosing by arrival time; as written, the design does not ensure that two teams' analyses are reconciled before publication.
D:In an engineering knowledge assistant, merging divergent work requires an authoritative step rather than choosing by arrival time, but full replacement risks losing unrelated code and formatting; therefore the separate requirement that an existing file receives a precise localized replacement would remain unsatisfied.
F:In an engineering knowledge assistant, merging divergent work requires an authoritative step rather than choosing by arrival time, but the independent mechanism for the other criterion is absent: surgical edits preserve unrelated lines and provide reviewable diffs; as written, the design does not ensure that an existing file receives a precise localized replacement.
Question 666
An architecture review of a document field validation service finds two missing guarantees: an uncertain high-impact decision is routed for review instead of automatic execution; a consequential claim receives independent human sign-off. What two-part change closes both gaps?
- proceed whenever the assistant's prose sounds certain; and accept model confidence as a substitute for authorization
- proceed whenever the assistant's prose sounds certain; and route high-impact or low-evidence results to an authorized reviewer
- set a clear evidence or confidence boundary and require review below it; and route high-impact or low-evidence results to an authorized reviewer
- set a clear evidence or confidence boundary and require review below it; and accept model confidence as a substitute for authorization
Explanation
Correct Answer
C:In a document field validation service, risk-weighted gates limit autonomous actions when support is inadequate; alongside this, human approval supplies accountability where a mistaken model result has material cost, satisfying the distinct requirements that an uncertain high-impact decision is routed for review instead of automatic execution and that a consequential claim receives independent human sign-off.
Incorrect Answers
A:In a document field validation service, fluent language is not a calibrated reliability measure and a model score does not confer business or legal authority; the design would fail both conditions: an uncertain high-impact decision is routed for review instead of automatic execution and a consequential claim receives independent human sign-off.
B:In a document field validation service, human approval supplies accountability where a mistaken model result has material cost, but fluent language is not a calibrated reliability measure; therefore the separate requirement that an uncertain high-impact decision is routed for review instead of automatic execution would remain unsatisfied.
D:In a document field validation service, risk-weighted gates limit autonomous actions when support is inadequate, but a model score does not confer business or legal authority; therefore the separate requirement that a consequential claim receives independent human sign-off would remain unsatisfied.
Question 667
An incident postmortem for a code-reference retrieval assistant requires controls so that specialists return comparable artifacts for aggregation and that a paused server-side tool loop can continue without losing work. Which paired remedy covers both failure modes?
- merge narrative responses by concatenating their text; and preserve the paused assistant response and resume the server-tool turn
- merge narrative responses by concatenating their text; and restart from a blank conversation whenever pause_turn occurs
- give subagents a common results contract including claim, evidence, and status; and preserve the paused assistant response and resume the server-tool turn
- give subagents a common results contract including claim, evidence, and status; and restart from a blank conversation whenever pause_turn occurs
Explanation
Correct Answer
C:In a code-reference retrieval assistant, a typed handoff lets the coordinator compare like fields and detect missing support; alongside this, pause_turn indicates an internal tool loop pause rather than a user-facing conclusion, satisfying the distinct requirements that specialists return comparable artifacts for aggregation and that a paused server-side tool loop can continue without losing work.
Incorrect Answers
A:In a code-reference retrieval assistant, pause_turn indicates an internal tool loop pause rather than a user-facing conclusion, but unstructured concatenation hides provenance and duplicate or conflicting claims; therefore the separate requirement that specialists return comparable artifacts for aggregation would remain unsatisfied.
B:In a code-reference retrieval assistant, unstructured concatenation hides provenance and duplicate or conflicting claims and dropping tool context can repeat work and omit intermediate evidence; the design would fail both conditions: specialists return comparable artifacts for aggregation and a paused server-side tool loop can continue without losing work.
D:In a code-reference retrieval assistant, a typed handoff lets the coordinator compare like fields and detect missing support, but dropping tool context can repeat work and omit intermediate evidence; therefore the separate requirement that a paused server-side tool loop can continue without losing work would remain unsatisfied.
Question 668
A readiness review for a large-codebase architecture explorer requires that multiple subagent outputs can be traced to their individual tasks; that a selector accepts only supported operation names; and that a destructive tool receives human authorization at execution time. Which THREE actions directly implement those distinct requirements? Choose THREE.
- declare an operation enum in the schema and reject values outside it in the tool handler
- collect a general approval at the start of the conversation
- interrupt or deny the tool until an authorization callback approves it
- ask the model to invent a suitable action string
- attach task IDs and source references to returned artifacts
- strip origin metadata from each response before merging
Explanation
Correct Answers
A:For a large-codebase architecture explorer, schema constraints guide valid selections and runtime validation blocks unsupported operation values, which directly establishes that a selector accepts only supported operation names; the separate requirement that a destructive tool receives human authorization at execution time is handled by a different action.
C:For a large-codebase architecture explorer, execution gates make the approval binding on the selected action and inputs, which directly establishes that a destructive tool receives human authorization at execution time; the separate requirement that multiple subagent outputs can be traced to their individual tasks is handled by a different action.
E:For a large-codebase architecture explorer, traceable results let the coordinator identify which delegation produced each claim, which directly establishes that multiple subagent outputs can be traced to their individual tasks; the separate requirement that a selector accepts only supported operation names is handled by a different action.
Incorrect Answers
B:For a large-codebase architecture explorer, blanket consent does not verify the later high-risk operation, so this option cannot establish that a destructive tool receives human authorization at execution time; correctly addressing multiple subagent outputs can be traced to their individual tasks elsewhere does not repair that deficiency.
D:For a large-codebase architecture explorer, free-form labels can name unimplemented operations, and a prompt is not an enforced allowlist, so this option cannot establish that a selector accepts only supported operation names; correctly addressing a destructive tool receives human authorization at execution time elsewhere does not repair that deficiency.
F:For a large-codebase architecture explorer, source-less aggregation impairs diagnosis and fact checking, so this option cannot establish that multiple subagent outputs can be traced to their individual tasks; correctly addressing a selector accepts only supported operation names elsewhere does not repair that deficiency.
Question 669
In a repository documentation project, the acceptance criteria require that only a nested service uses specialized implementation rules and that the agent executes tests and examines their exit status. Which paired implementation meets both?
- place a dedicated CLAUDE.md in the relevant directory; and inspect the test configuration only with Read and claim tests passed
- put all service-only details in the repository root instruction file; and use Bash with an approved test command
- put all service-only details in the repository root instruction file; and inspect the test configuration only with Read and claim tests passed
- place a dedicated CLAUDE.md in the relevant directory; and use Bash with an approved test command
Explanation
Correct Answer
D:In a repository documentation project, directory-specific context can refine instructions for work under that subtree; alongside this, the shell can run build tools and return execution feedback that Read cannot supply, satisfying the distinct requirements that only a nested service uses specialized implementation rules and that the agent executes tests and examines their exit status.
Incorrect Answers
A:In a repository documentation project, directory-specific context can refine instructions for work under that subtree, but reading configuration does not execute a test suite; therefore the separate requirement that the agent executes tests and examines their exit status would remain unsatisfied.
B:In a repository documentation project, the shell can run build tools and return execution feedback that Read cannot supply, but globalizing narrow rules adds irrelevant instructions to unrelated work; therefore the separate requirement that only a nested service uses specialized implementation rules would remain unsatisfied.
C:In a repository documentation project, globalizing narrow rules adds irrelevant instructions to unrelated work and reading configuration does not execute a test suite; the design would fail both conditions: only a nested service uses specialized implementation rules and the agent executes tests and examines their exit status.
Question 670
When designing a returns-and-billing assistant, the requirement is not satisfied unless an extractor always returns the identifiers required by downstream code and a support case missing an essential fact is not silently guessed. Which implementation addresses the two independent conditions?
- treat every identifier as an optional free-form note; and ask a focused clarifying question for the missing requirement
- treat every identifier as an optional free-form note; and invent a plausible value and continue automatically
- declare mandatory schema fields in the required list; and invent a plausible value and continue automatically
- declare mandatory schema fields in the required list; and ask a focused clarifying question for the missing requirement
Explanation
Correct Answer
D:In a returns-and-billing assistant, required properties prevent accidental omission of key attributes; alongside this, targeted clarification supplies the decision input before selecting an irreversible action, satisfying the distinct requirements that an extractor always returns the identifiers required by downstream code and that a support case missing an essential fact is not silently guessed.
Incorrect Answers
A:In a returns-and-billing assistant, targeted clarification supplies the decision input before selecting an irreversible action, but the consumer cannot reliably rely on a field that may be omitted; therefore the separate requirement that an extractor always returns the identifiers required by downstream code would remain unsatisfied.
B:In a returns-and-billing assistant, the consumer cannot reliably rely on a field that may be omitted and fabricated facts can cause the wrong account or policy to be applied; the design would fail both conditions: an extractor always returns the identifiers required by downstream code and a support case missing an essential fact is not silently guessed.
C:In a returns-and-billing assistant, required properties prevent accidental omission of key attributes, but fabricated facts can cause the wrong account or policy to be applied; therefore the separate requirement that a support case missing an essential fact is not silently guessed would remain unsatisfied.