Question 280
An internal compliance system tracks where evidence for decisions originated. An extraction result is copied into a report with no link to its document. Which option best meets the operational need?
- Record retrieval time and a content digest or archived excerpt where permitted
- Carry source ID, page, and extraction run identifier into the final record
- Compare source-of-truth ownership and freshness before selecting a status
- Mark quotations, extracted facts, and analyst inferences separately
- Attach the document version and effective date to the quoted passage
Explanation
Correct Answer
B:A provenance chain connects the published value to the observed document and the specific processing run that produced it.
Incorrect Answers
A:When a cited webpage changes after the original analysis was completed, this would be useful because web evidence is mutable; a stable reference explains what the system actually saw when reaching its original conclusion; this item instead calls for a way to retain lineage through transformations.
C:A newer timestamp is not always sufficient if one system is a derived cache; authority and synchronization matter; the alternative would make sense if two tools return different customer statuses from separate data stores. It does not directly fix the observed condition: an extraction result is copied into a report with no link to its document.
D:This measure fits a different trigger: a research report mixes direct quotations and model paraphrases without labeling them. In that setting, Separating evidence types prevents a model's synthesis from appearing to be a verbatim statement by an external authority; it does not resolve the present need to retain lineage through transformations.
E:Policy text changes over time; revision identity prevents a once-valid rule from being applied to the wrong period; the relevant precondition would be a policy interpretation quotes a paragraph but does not identify the policy revision. The question instead describes an extraction result is copied into a report with no link to its document.
Question 281
An engineering assistant writes migration plans for internal reviewers. A risk register mixes confirmed issues with guesses. What is the most appropriate design choice?
- Evaluate each alternative against the same security, latency, maintenance, and failure criteria
- Require each risk to be marked observed, inferred, or unknown with supporting evidence
- Request checkpoints, reversal conditions, and limitations of restoring state
- Require backward compatibility, staged rollout, and rollback considerations
Explanation
Correct Answer
B:Distinguishing established facts from hypotheses prevents speculative risk language from masquerading as a verified defect.
Incorrect Answers
A:When the team needs to compare two architecture options fairly, this would be useful because a shared rubric makes trade-offs comparable rather than giving one preferred option more favorable assessment dimensions; this item instead calls for a way to separate evidence status in output criteria.
C:This measure fits a different trigger: a reviewer cares about data migration reversibility. In that setting, Rollback feasibility must be assessed during planning, not assumed after the assistant recommends irreversible changes; it does not resolve the present need to separate evidence status in output criteria.
D:An explicit availability constraint changes feasible migration designs; it must be visible before the model proposes modifications; the alternative would make sense if a rollout has a strict zero-downtime requirement. It does not directly fix the observed condition: a risk register mixes confirmed issues with guesses.
Question 282
A Claude-based coding assistant generates patch-review records. The assistant proposes tests that do not run in the current build. Which response is most reliable?
- Run permitted commands and report any unavailable or failing test conditions
- Mark missing checks as blocked and do not assign a passing verdict
- Request targeted correction while retaining validated findings
- Check line ranges against the exact repository revision before accepting findings
Explanation
Correct Answer
A:Describing test commands does not prove they are executable or passing; CI evidence is required. For the stated case, the check is whether the system can execute suggested tests as a separate acceptance gate; this approach supplies that control.
Incorrect Answers
B:When the model returns a plausible patch summary even though required static checks were skipped, this would be useful because a fluent summary cannot turn skipped checks into evidence; state should reflect actual execution results; this item instead calls for a way to execute suggested tests as a separate acceptance gate.
C:This measure fits a different trigger: a correction retry rewrites already verified findings unnecessarily. In that setting, Narrow repair reduces regression risk and prevents fresh hallucinations in sections that already passed checking; it does not resolve the present need to execute suggested tests as a separate acceptance gate.
D:Schema integers can still point outside actual files; source-aware validation prevents misleading review annotations; the relevant precondition would be a line number in a valid review object exceeds the length of the referenced file. The question instead describes the assistant proposes tests that do not run in the current build.
Question 283
A multi-agent study summarizes academic papers and compares their claims. The final report contains a claim that no specialist can support. What should the team do to resolve this particular gap?
- Cache safe read-only metadata results keyed by source and version
- Block publication of claims lacking linked evidence or flag them for review
- Create a conflict record containing both definitions and their source contexts
- Make methods extraction a prerequisite for the evaluation stage
Explanation
Correct Answer
B:A synthesis should not invent authority; a claim-to-source mapping makes unsupported assertions detectable. Here it serves the explicit need to exclude or qualify unsupported claims without adding a separate unverified inference.
Incorrect Answers
A:When the research workload includes ten identical metadata lookups, this would be useful because reusing verified immutable metadata reduces repeated calls without treating dynamic or personalized search results as permanently correct; this item instead calls for a way to exclude or qualify unsupported claims.
C:This measure fits a different trigger: a review worker finds two incompatible definitions of the same metric. In that setting, Numerical reconciliation is not meaningful if the measurements refer to different constructs; definitions must be resolved first; it does not resolve the present need to exclude or qualify unsupported claims.
D:A conclusion without methods context can overstate reliability; dependency ordering preserves critical evaluation evidence; the alternative would make sense if a paper has a methods section that must be checked before its conclusion is summarized. It does not directly fix the observed condition: the final report contains a claim that no specialist can support.
Question 284
An unfamiliar codebase requires careful changes through Claude Code. The team wants proof that a bug is fixed, not just a plausible description. Which control is most directly relevant?
- Search the repository and read referenced files before proposing modifications
- Change the generator input and regenerate outputs using the supported workflow
- Capture the original reproducible failure and the verified result after the patch
- Mark the integration check blocked and provide the exact human-run requirement
Explanation
Correct Answer
C:A before-and-after test provides stronger evidence of correction than a narrative assertion alone. For the stated case, the check is whether the system can demonstrate the failing-then-passing test sequence; this approach supplies that control.
Incorrect Answers
A:File evidence prevents made-up dependencies and directs changes to the real code organization; the relevant precondition would be Claude starts inventing module names instead of inspecting the repository. The question instead describes the team wants proof that a bug is fixed, not just a plausible description.
B:This measure fits a different trigger: the assistant modified generated files that will be overwritten by builds. In that setting, Derived artifacts are not authoritative; editing them directly is fragile and hides the persistent source of the behavior; it does not resolve the present need to demonstrate the failing-then-passing test sequence.
D:When the acceptance test depends on a credential unavailable in the agent workspace, this would be useful because an unrun test must not be reported as passing; explicit blocked status lets the maintainer complete verification safely; this item instead calls for a way to demonstrate the failing-then-passing test sequence.
Question 285
A development agent explores competing fixes across multiple sessions. One experiment altered local files that another experiment must not see. How should the workflow be adjusted?
- Use separate worktrees or sandboxes for mutable branch experiments
- Fork a common session context for each experiment
- Preserve goals, constraints, decisions, open issues, and source references while discarding noise
- Resume the relevant session and inspect the recorded pending work
- Rerun relevant tests on the current revision before accepting success
Explanation
Correct Answer
A:Forked dialogue history alone does not isolate disk writes; workspace separation prevents cross-experiment contamination.
Incorrect Answers
B:Branching keeps baseline assumptions comparable, whereas serially editing one conversation blends conclusions; the alternative would make sense if two proposed fixes need separate experiments from the same baseline. It does not directly fix the observed condition: one experiment altered local files that another experiment must not see.
C:When a session has grown so large that prior assumptions hide the active requirement, this would be useful because structured compaction retains what is needed for continuation and reduces irrelevant history without implying forgotten facts are still available; this item instead calls for a way to isolate filesystem state as well as context.
D:Continuing the actual conversation preserves intent and decisions, although the current code must still be validated; the relevant precondition would be a developer needs to continue yesterday's unfinished refactor. The question instead describes one experiment altered local files that another experiment must not see.
E:This measure fits a different trigger: a resumed worker reports success based on a cached test from last week. In that setting, Tests describe a specific code state; a prior green result cannot establish correctness for a changed repository; it does not resolve the present need to isolate filesystem state as well as context.
Question 286
A support team exposes several similar billing tools through an MCP server. An account lookup can handle only active accounts, which is absent from its description. Which implementation decision is the best fit?
- Declare meaningful JSON Schema types, required fields, and relevant constraints
- Differentiate names, descriptions, required inputs, and result semantics by real capability
- Return a concise typed result with the required fields and stable meanings
- Describe the active-account limitation and expected error behavior in the tool contract
Explanation
Correct Answer
D:A precondition affects whether the model should choose the tool; hiding it leads to avoidable failures and misleading answers.
Incorrect Answers
A:Clear argument schemas improve generated calls and let the host reject malformed or missing identifiers before sensitive execution; the alternative would make sense if a refund tool accepts order IDs but its schema describes an untyped free-form string. It does not directly fix the observed condition: an account lookup can handle only active accounts, which is absent from its description.
B:This measure fits a different trigger: two tools have nearly identical names and descriptions despite different output formats. In that setting, Precise boundaries are more effective than extra prompt reminders when Claude must select between overlapping functions; it does not resolve the present need to document a selection precondition.
C:When a tool returns a large raw object although callers need just three fields, this would be useful because a narrow result format saves context and keeps downstream reasoning focused on validated data rather than incidental service metadata; this item instead calls for a way to document a selection precondition.
Question 287
A Claude Agent SDK workflow delegates repository investigations to subagents. Two independent test modules can be investigated separately. What should the architect specify in the design?
- Launch the independent module inquiries in parallel and merge results later
- Give the Task-oriented specialist a concrete goal, scope, and expected handoff
- Restrict the worker's allowedTools to safe inspection operations
- Run the investigation in a subagent and return a concise evidence-based summary
Explanation
Correct Answer
A:Concurrent subagents can reduce wall-clock time when their files and decisions do not have data dependencies. This matches the workflow goal to use suitable subagent concurrency rather than introducing an unrelated action.
Incorrect Answers
B:This measure fits a different trigger: a parent agent needs a specialist to take a bounded action rather than just offer advice. In that setting, Structured delegated tasks clarify which worker owns the operation and how its outcome will return to the parent; it does not resolve the present need to use suitable subagent concurrency.
C:Tool allowlists define an execution boundary that remains effective even when the worker's text suggests a dangerous action; the relevant precondition would be a worker should be able to inspect code but not run destructive shell commands. The question instead describes two independent test modules can be investigated separately.
D:An isolated worker context avoids flooding the coordinator with irrelevant file reads while preserving the actionable findings; the alternative would make sense if a specialist's large file-search transcript is not useful to the parent. It does not directly fix the observed condition: two independent test modules can be investigated separately.
Question 288
A moderation assistant must apply a policy rubric consistently. The prompt includes examples with conflicting labels for the same situation. Which change most directly addresses this condition?
- Add counterexamples that show when the rule does not apply
- Keep a small set of examples that represent distinct failure modes
- Adjudicate contradictory samples and keep one coherent set
- Include a minimal pair showing how the exception changes the judgment
Explanation
Correct Answer
C:Inconsistent demonstrations make the target behavior ill-defined and can be worse than a simpler instruction. For the stated case, the check is whether the system can repair label consistency before scaling; this approach supplies that control.
Incorrect Answers
A:One-sided examples bias the output; contrastive cases teach the difference between related categories; the alternative would make sense if all examples in the prompt are easy positive cases. It does not directly fix the observed condition: the prompt includes examples with conflicting labels for the same situation.
B:When the few-shot examples are longer than the actual task input, this would be useful because verbose demonstrations consume context; coverage of tricky distinctions matters more than repeating easy examples; this item instead calls for a way to repair label consistency before scaling.
D:This measure fits a different trigger: a new exception is introduced to the policy. In that setting, Near-identical cases with different required outcomes can clarify the actual boundary introduced by a rule change; it does not resolve the present need to repair label consistency before scaling.
Question 289
A research MCP server uses structured errors and pagination. An article is accessible but its content type is not supported by the parser. What is the most defensible next step?
- Return successful index results plus per-source error statuses
- Expose a sanitized error code and trace ID while logging restricted details internally
- Return a timeout category with correlation ID and bounded retry guidance
- Return an unsupported-format status with the detected media type
Explanation
Correct Answer
D:A format mismatch needs a different parser or workflow, not repeated identical calls to an unavailable feature. Here it serves the explicit need to distinguish unsupported data from service errors without adding a separate unverified inference.
Incorrect Answers
A:This measure fits a different trigger: only one of five federated indexes fails while others succeed. In that setting, A mixed outcome should retain usable evidence while informing the coordinator that coverage is incomplete; it does not resolve the present need to distinguish unsupported data from service errors.
B:When a request fails with an internal stack trace containing server paths, this would be useful because structured diagnostics allow troubleshooting without dumping sensitive service internals into a language-model conversation; this item instead calls for a way to distinguish unsupported data from service errors.
C:Time limits protect system capacity; reporting an empty search result would conceal that no reliable search completed; the alternative would make sense if an upstream search endpoint times out after several seconds. It does not directly fix the observed condition: an article is accessible but its content type is not supported by the parser.
Question 290
A research agent must explore conflicting hypotheses with limited resources. An answer requires independently checking both supportive and contradictory evidence. Which approach satisfies the stated constraint?
- Stop additional investigation once evidence suffices and obligations are satisfied
- Track completed task intents and reuse valid existing findings
- Allow a bounded agent loop to generate follow-ups from validated findings
- Enumerate documents and allocate bounded extraction tasks to each
- Assign explicit confirmation and challenge tasks before the final synthesis
Explanation
Correct Answer
E:Separate search objectives reduce confirmation bias and give the coordinator grounds to compare conflicting material.
Incorrect Answers
A:When an expensive analysis path has low expected value after a reliable answer is found, this would be useful because a clear sufficiency condition avoids waste while preserving any mandatory validation or approval steps; this item instead calls for a way to decompose by opposing evidentiary roles.
B:This measure fits a different trigger: the coordinator keeps asking new workers for the same question. In that setting, Deduplicating work prevents an apparent multi-agent design from simply repeating the same investigation; it does not resolve the present need to decompose by opposing evidentiary roles.
C:Uncertain discovery benefits from adaptive next steps; the budget and tools still require host control; the alternative would make sense if an initial article may reveal unknown organizations that require follow-up queries. It does not directly fix the observed condition: an answer requires independently checking both supportive and contradictory evidence.
D:A finite inventory can be processed predictably without an autonomous planner deciding what to look at next; the relevant precondition would be the search space has a fixed list of ten known documents. The question instead describes an answer requires independently checking both supportive and contradictory evidence.