Question 770
In a technical review of a developer productivity assistant, the architect must address the following constraints together: the model can use a tool result in its next decision; conflicting subagent findings produce one accountable interpretation. Which change set is most defensible?
- put the tool output only in application logs; and have the coordinator compare claims and sources before synthesis
- append the assistant tool-use response and matching user tool_result blocks to history; and let the fastest subagent override other findings automatically
- implement only append the assistant tool-use response and matching user tool_result blocks to history; make no corresponding change for the second requirement
- append the assistant tool-use response and matching user tool_result blocks to history; and have the coordinator compare claims and sources before synthesis
- put the tool output only in application logs; and let the fastest subagent override other findings automatically
Explanation
Correct Answer
D:In a developer productivity assistant, Claude needs the prior request and the execution output in the conversation for the next inference; alongside this, central reconciliation can distinguish evidence disagreement from a formatting difference, satisfying the distinct requirements that the model can use a tool result in its next decision and that conflicting subagent findings produce one accountable interpretation.
Incorrect Answers
A:In a developer productivity assistant, central reconciliation can distinguish evidence disagreement from a formatting difference, but unseen logs cannot inform the model's next decision; therefore the separate requirement that the model can use a tool result in its next decision would remain unsatisfied.
B:In a developer productivity assistant, Claude needs the prior request and the execution output in the conversation for the next inference, but completion order is not a sound evidence-ranking rule; therefore the separate requirement that conflicting subagent findings produce one accountable interpretation would remain unsatisfied.
C:In a developer productivity assistant, Claude needs the prior request and the execution output in the conversation for the next inference, but the independent mechanism for the other criterion is absent: central reconciliation can distinguish evidence disagreement from a formatting difference; as written, the design does not ensure that conflicting subagent findings produce one accountable interpretation.
E:In a developer productivity assistant, unseen logs cannot inform the model's next decision and completion order is not a sound evidence-ranking rule; the design would fail both conditions: the model can use a tool result in its next decision and conflicting subagent findings produce one accountable interpretation.
Question 771
In a technical review of a warranty claims assistant, the architect must address the following constraints together: tool-call logging does not modify a successful tool result; context reduction does not erase unresolved obligations. Which change set is most defensible?
- rewrite every tool result as a free-form paraphrase; and assume the agent will recall every open action from memory
- use an observation-only hook for telemetry when rewriting is unnecessary; and track pending approvals and required next steps in durable task state
- use an observation-only hook for telemetry when rewriting is unnecessary; and assume the agent will recall every open action from memory
- rewrite every tool result as a free-form paraphrase; and track pending approvals and required next steps in durable task state
- implement only use an observation-only hook for telemetry when rewriting is unnecessary; make no corresponding change for the second requirement
Explanation
Correct Answer
B:In a warranty claims assistant, separating instrumentation from mutation reduces the risk of corrupting evidence; alongside this, persisted obligations survive when natural-language history is shortened, satisfying the distinct requirements that tool-call logging does not modify a successful tool result and that context reduction does not erase unresolved obligations.
Incorrect Answers
A:In a warranty claims assistant, summaries can discard fields that subsequent steps require and window truncation can drop an outstanding approval or escalation; the design would fail both conditions: tool-call logging does not modify a successful tool result and context reduction does not erase unresolved obligations.
C:In a warranty claims assistant, separating instrumentation from mutation reduces the risk of corrupting evidence, but window truncation can drop an outstanding approval or escalation; therefore the separate requirement that context reduction does not erase unresolved obligations would remain unsatisfied.
D:In a warranty claims assistant, persisted obligations survive when natural-language history is shortened, but summaries can discard fields that subsequent steps require; therefore the separate requirement that tool-call logging does not modify a successful tool result would remain unsatisfied.
E:In a warranty claims assistant, separating instrumentation from mutation reduces the risk of corrupting evidence, but the independent mechanism for the other criterion is absent: persisted obligations survive when natural-language history is shortened; as written, the design does not ensure that context reduction does not erase unresolved obligations.
Question 772
An incident postmortem for an expense receipt capture system requires controls so that each tool interface has a single clear operation boundary and that tool responses are associated with the exact invocations that produced them. Which paired remedy covers both failure modes?
- separate read-only lookup from side-effecting update interfaces; and return tool results with newly invented identifiers
- separate read-only lookup from side-effecting update interfaces; and return tool_result blocks using the original tool_use_id values
- use a single universal tool that infers read versus write from prose; and return tool results with newly invented identifiers
- use a single universal tool that infers read versus write from prose; and return tool_result blocks using the original tool_use_id values
- implement only separate read-only lookup from side-effecting update interfaces; make no corresponding change for the second requirement
Explanation
Correct Answer
B:In an expense receipt capture system, clear operation boundaries support least privilege and understandable approvals; alongside this, correlation identifiers bind asynchronous or parallel results to the right requests, satisfying the distinct requirements that each tool interface has a single clear operation boundary and that tool responses are associated with the exact invocations that produced them.
Incorrect Answers
A:In an expense receipt capture system, clear operation boundaries support least privilege and understandable approvals, but the model cannot reliably associate a result with an outstanding call; therefore the separate requirement that tool responses are associated with the exact invocations that produced them would remain unsatisfied.
C:In an expense receipt capture system, overloading hides effects and makes execution review harder and the model cannot reliably associate a result with an outstanding call; the design would fail both conditions: each tool interface has a single clear operation boundary and tool responses are associated with the exact invocations that produced them.
D:In an expense receipt capture system, correlation identifiers bind asynchronous or parallel results to the right requests, but overloading hides effects and makes execution review harder; therefore the separate requirement that each tool interface has a single clear operation boundary would remain unsatisfied.
E:In an expense receipt capture system, clear operation boundaries support least privilege and understandable approvals, but the independent mechanism for the other criterion is absent: correlation identifiers bind asynchronous or parallel results to the right requests; as written, the design does not ensure that tool responses are associated with the exact invocations that produced them.
Question 773
An incident postmortem for a code policy conformance job requires controls so that project guidance fits within a limited context budget and that documentation-only conventions do not dilute code-editing tasks. Which paired remedy covers both failure modes?
- paste complete design manuals into the top-level CLAUDE.md; and place all documentation writing rules in a global user file
- paste complete design manuals into the top-level CLAUDE.md; and scope documentation style rules to matching documentation paths
- keep the top-level instruction set concise and move specialized details to scoped files; and scope documentation style rules to matching documentation paths
- keep the top-level instruction set concise and move specialized details to scoped files; and place all documentation writing rules in a global user file
Explanation
Correct Answer
C:In a code policy conformance job, focused guidance preserves useful context and reduces instruction competition; alongside this, conditional loading keeps instruction context relevant to the file being worked on, satisfying the distinct requirements that project guidance fits within a limited context budget and that documentation-only conventions do not dilute code-editing tasks.
Incorrect Answers
A:In a code policy conformance job, indiscriminate loading spends tokens on facts unrelated to the active task and those conventions will follow unrelated code tasks; the design would fail both conditions: project guidance fits within a limited context budget and documentation-only conventions do not dilute code-editing tasks.
B:In a code policy conformance job, conditional loading keeps instruction context relevant to the file being worked on, but indiscriminate loading spends tokens on facts unrelated to the active task; therefore the separate requirement that project guidance fits within a limited context budget would remain unsatisfied.
D:In a code policy conformance job, focused guidance preserves useful context and reduces instruction competition, but those conventions will follow unrelated code tasks; therefore the separate requirement that documentation-only conventions do not dilute code-editing tasks would remain unsatisfied.
Question 774
For a symbol-search assistant, a reviewer identifies two independent obligations: the developer inspects a suspicious implementation in context and conflicting guidance from different files is detected before rollout. Which proposal should be approved?
- read targeted code ranges and nearby dependencies; and audit overlapping CLAUDE.md files and resolve instruction contradictions
- replace source inspection with a guessed architecture overview; and assume the model will infer which contradictory rule to follow
- replace source inspection with a guessed architecture overview; and audit overlapping CLAUDE.md files and resolve instruction contradictions
- read targeted code ranges and nearby dependencies; and assume the model will infer which contradictory rule to follow
- implement only read targeted code ranges and nearby dependencies; make no corresponding change for the second requirement
Explanation
Correct Answer
A:In a symbol-search assistant, focused inspection preserves exact implementation detail without loading the whole codebase; alongside this, clear ownership and scope prevent two directions from competing in the same coding task, satisfying the distinct requirements that the developer inspects a suspicious implementation in context and that conflicting guidance from different files is detected before rollout.
Incorrect Answers
B:In a symbol-search assistant, a general overview cannot verify the specific logic being debugged and unresolved conflicts create inconsistent output across directories; the design would fail both conditions: the developer inspects a suspicious implementation in context and conflicting guidance from different files is detected before rollout.
C:In a symbol-search assistant, clear ownership and scope prevent two directions from competing in the same coding task, but a general overview cannot verify the specific logic being debugged; therefore the separate requirement that the developer inspects a suspicious implementation in context would remain unsatisfied.
D:In a symbol-search assistant, focused inspection preserves exact implementation detail without loading the whole codebase, but unresolved conflicts create inconsistent output across directories; therefore the separate requirement that conflicting guidance from different files is detected before rollout would remain unsatisfied.
E:In a symbol-search assistant, focused inspection preserves exact implementation detail without loading the whole codebase, but the independent mechanism for the other criterion is absent: clear ownership and scope prevent two directions from competing in the same coding task; as written, the design does not ensure that conflicting guidance from different files is detected before rollout.
Question 775
An overnight claims-form extraction pipeline must process thousands of independent documents, and every persisted case status must belong to an approved enumeration. Which combined approach handles both workload scale and status validation?
- accept any model-generated status description; and make every document a synchronous request inside the user interaction
- enforce the approved status enum with structured output and validate it in the consumer; and make every document a synchronous request inside the user interaction
- enforce the approved status enum with structured output and validate it in the consumer; and use asynchronous Message Batches processing for the bulk workload
- implement only enforce the approved status enum with structured output and validate it in the consumer; make no corresponding change for the second requirement
- accept any model-generated status description; and use asynchronous Message Batches processing for the bulk workload
Explanation
Correct Answer
C:In a claims form processing assistant, a constrained schema expresses the allowed labels and downstream validation rejects out-of-contract values; alongside this, batching suits non-blocking independent requests and can reduce processing cost, satisfying the distinct requirements that a status field can contain only approved business states and that thousands of independent documents can finish without interactive response deadlines.
Incorrect Answers
A:In a claims form processing assistant, free-form status phrases can violate a fixed business-state contract and serial immediate calls add latency where no caller is waiting; the design would fail both conditions: a status field can contain only approved business states and thousands of independent documents can finish without interactive response deadlines.
B:In a claims form processing assistant, a constrained schema expresses the allowed labels and downstream validation rejects out-of-contract values, but serial immediate calls add latency where no caller is waiting; therefore the separate requirement that thousands of independent documents can finish without interactive response deadlines would remain unsatisfied.
D:In a claims form processing assistant, a constrained schema expresses the allowed labels and downstream validation rejects out-of-contract values, but the independent mechanism for the other criterion is absent: batching suits non-blocking independent requests and can reduce processing cost; as written, the design does not ensure that thousands of independent documents can finish without interactive response deadlines.
E:In a claims form processing assistant, batching suits non-blocking independent requests and can reduce processing cost, but free-form status phrases can violate a fixed business-state contract; therefore the separate requirement that a status field can contain only approved business states would remain unsatisfied.
Question 776
The release gate for a large-codebase architecture explorer requires that a code investigation finds where a function is referenced; it also requires that all contributors receive the repository's stable engineering conventions. Which TWO actions meet the specified checks? Choose TWO.
- put team-wide rules only in one developer's user-level instructions
- ask for broad prose summaries of every package first
- put concise shared guidance in the repository root CLAUDE.md
- use Grep to locate occurrences before opening focused files
Explanation
Correct Answers
C:For a large-codebase architecture explorer, project instructions travel with the codebase and establish common architecture and build constraints, which directly establishes that all contributors receive the repository's stable engineering conventions; the separate requirement that a code investigation finds where a function is referenced is handled by a different action.
D:For a large-codebase architecture explorer, symbol searching narrows the evidence to files that mention the target, which directly establishes that a code investigation finds where a function is referenced; the separate requirement that all contributors receive the repository's stable engineering conventions is handled by a different action.
Incorrect Answers
A:For a large-codebase architecture explorer, personal files do not distribute the intended conventions to other contributors, so this option cannot establish that all contributors receive the repository's stable engineering conventions; correctly addressing a code investigation finds where a function is referenced elsewhere does not repair that deficiency.
B:For a large-codebase architecture explorer, high-level summaries can miss the exact call site, so this option cannot establish that a code investigation finds where a function is referenced; correctly addressing all contributors receive the repository's stable engineering conventions elsewhere does not repair that deficiency.
Question 777
An architecture review of a symbol-search assistant finds two missing guarantees: a failed high-stakes step prevents a false final success signal; independent research branches can finish without waiting on one another. What two-part change closes both gaps?
- ignore errors after the first successful agent completion; and chain every research agent in a fixed serial pipeline
- enforce required-stage completion before publishing a success status; and chain every research agent in a fixed serial pipeline
- enforce required-stage completion before publishing a success status; and dispatch the independent branches concurrently through the coordinator
- ignore errors after the first successful agent completion; and dispatch the independent branches concurrently through the coordinator
Explanation
Correct Answer
C:In a symbol-search assistant, a final gate must account for every mandatory stage rather than aggregate optimistic text; alongside this, parallel delegation reduces latency when individual investigations do not consume each other's outputs, satisfying the distinct requirements that a failed high-stakes step prevents a false final success signal and that independent research branches can finish without waiting on one another.
Incorrect Answers
A:In a symbol-search assistant, later failures can invalidate the claimed result and unnecessary sequential handoffs increase latency for independent subtasks; the design would fail both conditions: a failed high-stakes step prevents a false final success signal and independent research branches can finish without waiting on one another.
B:In a symbol-search assistant, a final gate must account for every mandatory stage rather than aggregate optimistic text, but unnecessary sequential handoffs increase latency for independent subtasks; therefore the separate requirement that independent research branches can finish without waiting on one another would remain unsatisfied.
D:In a symbol-search assistant, parallel delegation reduces latency when individual investigations do not consume each other's outputs, but later failures can invalidate the claimed result; therefore the separate requirement that a failed high-stakes step prevents a false final success signal would remain unsatisfied.
Question 778
An incident postmortem for a deployment-readiness verification process requires controls so that a coding agent inspects a known source file without changing it and that ambiguous requirements are resolved before a destructive migration. Which paired remedy covers both failure modes?
- use Read on the identified file; and make assumptions and migrate production files immediately
- run arbitrary shell commands to modify the file during inspection; and clarify scope and acceptance criteria while planning
- use Read on the identified file; and clarify scope and acceptance criteria while planning
- run arbitrary shell commands to modify the file during inspection; and make assumptions and migrate production files immediately
Explanation
Correct Answer
C:In a deployment-readiness verification process, file reading provides exact content for analysis without executing the project; alongside this, a reviewed plan should resolve unknown targets and rollback expectations before writes, satisfying the distinct requirements that a coding agent inspects a known source file without changing it and that ambiguous requirements are resolved before a destructive migration.
Incorrect Answers
A:In a deployment-readiness verification process, file reading provides exact content for analysis without executing the project, but guessed intent can create costly irreversible changes; therefore the separate requirement that ambiguous requirements are resolved before a destructive migration would remain unsatisfied.
B:In a deployment-readiness verification process, a reviewed plan should resolve unknown targets and rollback expectations before writes, but mutation is unnecessary for a read-only code review; therefore the separate requirement that a coding agent inspects a known source file without changing it would remain unsatisfied.
D:In a deployment-readiness verification process, mutation is unnecessary for a read-only code review and guessed intent can create costly irreversible changes; the design would fail both conditions: a coding agent inspects a known source file without changing it and ambiguous requirements are resolved before a destructive migration.
Question 779
Tests for a release-triage developer agent will pass only if a large research job combines specialized findings into one answer and if a delegated reviewer has no production write capability. Which proposed design directly satisfies both checks?
- have all subagents independently publish the final answer; and give the reviewer every administrative tool and ask for restraint
- have all subagents independently publish the final answer; and provide only the read-oriented tools required by that reviewer
- use a coordinator that delegates focused tasks and owns final synthesis; and provide only the read-oriented tools required by that reviewer
- use a coordinator that delegates focused tasks and owns final synthesis; and give the reviewer every administrative tool and ask for restraint
Explanation
Correct Answer
C:In a release-triage developer agent, the hub can integrate distinct subagent outputs and resolve overlap without peer-to-peer chatter; alongside this, agent-scoped tool exposure prevents a write path in a read-only role, satisfying the distinct requirements that a large research job combines specialized findings into one answer and that a delegated reviewer has no production write capability.
Incorrect Answers
A:In a release-triage developer agent, independent final outputs offer no single point to reconcile contradictions and prompt restraint is weaker than withholding execution capability; the design would fail both conditions: a large research job combines specialized findings into one answer and a delegated reviewer has no production write capability.
B:In a release-triage developer agent, agent-scoped tool exposure prevents a write path in a read-only role, but independent final outputs offer no single point to reconcile contradictions; therefore the separate requirement that a large research job combines specialized findings into one answer would remain unsatisfied.
D:In a release-triage developer agent, the hub can integrate distinct subagent outputs and resolve overlap without peer-to-peer chatter, but prompt restraint is weaker than withholding execution capability; therefore the separate requirement that a delegated reviewer has no production write capability would remain unsatisfied.
Question 780
While preparing a nightly code quality job for deployment, engineers must ensure that a developer returns to an interrupted session with prior reasoning; they must also ensure that separate code modules can be inspected in parallel. Which combination best satisfies those requirements?
- start an unrelated new session and assume it knows past actions; and spawn independent specialists in the same delegation stage
- resume the stored session by its identifier; and spawn independent specialists in the same delegation stage
- resume the stored session by its identifier; and make the second specialist wait for the first even though their modules are independent
- start an unrelated new session and assume it knows past actions; and make the second specialist wait for the first even though their modules are independent
Explanation
Correct Answer
B:In a nightly code quality job, session resumption retains the recorded conversation and avoids rebuilding the entire task from scratch; alongside this, concurrent subagents can examine non-overlapping modules before the coordinator merges their reports, satisfying the distinct requirements that a developer returns to an interrupted session with prior reasoning and that separate code modules can be inspected in parallel.
Incorrect Answers
A:In a nightly code quality job, concurrent subagents can examine non-overlapping modules before the coordinator merges their reports, but a fresh context lacks the prior interaction unless the facts are deliberately supplied; therefore the separate requirement that a developer returns to an interrupted session with prior reasoning would remain unsatisfied.
C:In a nightly code quality job, session resumption retains the recorded conversation and avoids rebuilding the entire task from scratch, but serialization wastes time without supplying a dependency; therefore the separate requirement that separate code modules can be inspected in parallel would remain unsatisfied.
D:In a nightly code quality job, a fresh context lacks the prior interaction unless the facts are deliberately supplied and serialization wastes time without supplying a dependency; the design would fail both conditions: a developer returns to an interrupted session with prior reasoning and separate code modules can be inspected in parallel.