Question 490
In a release-triage developer agent, the immediate design requirement is that the assistant locates files containing a particular symbol or string. Which change most directly establishes that behavior?
- use Write for file creation
- scan only filenames with Glob
- use Read on the identified file
- use Grep to search file contents
Explanation
Correct Answer
D:For a release-triage developer agent, text search locates occurrences efficiently before focused reading; this directly establishes that the assistant locates files containing a particular symbol or string instead of relying on an unrelated workflow control.
Incorrect Answers
A:Although writing a new artifact is appropriate when no existing target needs surgical modification, that only supports the distinct goal that a new source file is created from supplied content, not the requirement that the assistant locates files containing a particular symbol or string in a release-triage developer agent.
B:In a release-triage developer agent, path matching does not inspect symbol occurrences inside files; as a result, the stated requirement that the assistant locates files containing a particular symbol or string would still not be met.
C:Although file reading provides exact content for analysis without executing the project, that only supports the distinct goal that a coding agent inspects a known source file without changing it, not the requirement that the assistant locates files containing a particular symbol or string in a release-triage developer agent.
Question 491
In an internal knowledge research system, the immediate design requirement is that specialized agents work with only relevant permissions. Which change most directly establishes that behavior?
- define each subagent role and expose the minimum matching tools
- give every subagent the coordinator's complete tool inventory
- pass the preceding artifact identifier and a focused summary to the next agent
- specify the question, allowed sources, expected output, and stopping criterion
Explanation
Correct Answer
A:For an internal knowledge research system, tool scoping shrinks both decision confusion and the blast radius of delegated actions; this directly establishes that specialized agents work with only relevant permissions instead of relying on an unrelated workflow control.
Incorrect Answers
B:In an internal knowledge research system, unneeded write or network tools expand risk without improving the specialist task; as a result, the stated requirement that specialized agents work with only relevant permissions would still not be met.
C:Although explicit handoff preserves the specific result that the dependent stage must process, that only supports the distinct goal that a downstream subagent acts on concrete upstream results, not the requirement that specialized agents work with only relevant permissions in an internal knowledge research system.
D:Although a task contract prevents specialist exploration from swallowing the coordinator's larger goal, that only supports the distinct goal that delegated investigations stay within bounded scope, not the requirement that specialized agents work with only relevant permissions in an internal knowledge research system.
Question 492
Production readiness for a customer entitlement assistant depends on ensuring that invalid parameters lead to a clear correction rather than a blind retry and that out-of-policy exceptions receive an authorized decision. Select the most appropriate combined approach.
- hide the invalid field behind a generic internal-error message; and let the agent approve an exception whenever its wording sounds confident
- return a non-retryable validation error with the field and expected constraint; and route the exception to a human approval step before execution
- return a non-retryable validation error with the field and expected constraint; and let the agent approve an exception whenever its wording sounds confident
- hide the invalid field behind a generic internal-error message; and route the exception to a human approval step before execution
Explanation
Correct Answer
B:In a customer entitlement assistant, field-specific feedback lets the caller fix the input before resubmission; alongside this, approval is a workflow boundary separate from language-model confidence, satisfying the distinct requirements that invalid parameters lead to a clear correction rather than a blind retry and that out-of-policy exceptions receive an authorized decision.
Incorrect Answers
A:In a customer entitlement assistant, the client lacks the information needed to repair the request and model confidence is not delegated business authority; the design would fail both conditions: invalid parameters lead to a clear correction rather than a blind retry and out-of-policy exceptions receive an authorized decision.
C:In a customer entitlement assistant, field-specific feedback lets the caller fix the input before resubmission, but model confidence is not delegated business authority; therefore the separate requirement that out-of-policy exceptions receive an authorized decision would remain unsatisfied.
D:In a customer entitlement assistant, approval is a workflow boundary separate from language-model confidence, but the client lacks the information needed to repair the request; therefore the separate requirement that invalid parameters lead to a clear correction rather than a blind retry would remain unsatisfied.
Question 493
In a release-triage developer agent, the design must ensure that a team shares a repository-specific MCP server configuration and that delegated investigations stay within bounded scope. Choose TWO independent changes that directly meet these two requirements.
- commit a reviewed project .mcp.json without credentials
- configure each contributor's personal MCP registry manually
- ask every subagent to investigate everything it can find
- specify the question, allowed sources, expected output, and stopping criterion
Explanation
Correct Answers
A:For a release-triage developer agent, project-scoped configuration can be distributed with the repository while keeping secrets external, which directly establishes that a team shares a repository-specific MCP server configuration; the separate requirement that delegated investigations stay within bounded scope is handled by a different action.
D:For a release-triage developer agent, a task contract prevents specialist exploration from swallowing the coordinator's larger goal, which directly establishes that delegated investigations stay within bounded scope; the separate requirement that a team shares a repository-specific MCP server configuration is handled by a different action.
Incorrect Answers
B:For a release-triage developer agent, local personal settings do not create a reproducible team project setup, so this option cannot establish that a team shares a repository-specific MCP server configuration; correctly addressing delegated investigations stay within bounded scope elsewhere does not repair that deficiency.
C:For a release-triage developer agent, unbounded exploration dilutes signal and inflates context and latency, so this option cannot establish that delegated investigations stay within bounded scope; correctly addressing a team shares a repository-specific MCP server configuration elsewhere does not repair that deficiency.
Question 494
Tests for a vendor onboarding document pipeline will pass only if a classifier avoids marking uncertain cases as confirmed violations and if source identity survives multi-agent research handoffs. Which proposed design directly satisfies both checks?
- tell the model to flag anything that sounds suspicious; and strip all citation fields before sending the writer its inputs
- tell the model to flag anything that sounds suspicious; and carry citation identifiers and retrieval context through synthesis
- define explicit positive evidence and an abstain or review category; and strip all citation fields before sending the writer its inputs
- define explicit positive evidence and an abstain or review category; and carry citation identifiers and retrieval context through synthesis
Explanation
Correct Answer
D:In a vendor onboarding document pipeline, clear decision thresholds separate direct evidence from weak similarity; alongside this, provenance metadata must remain attached when intermediate summaries are merged, satisfying the distinct requirements that a classifier avoids marking uncertain cases as confirmed violations and that source identity survives multi-agent research handoffs.
Incorrect Answers
A:In a vendor onboarding document pipeline, vague semantic similarity raises false positives and the writer cannot reliably reconstruct which source supports each conclusion; the design would fail both conditions: a classifier avoids marking uncertain cases as confirmed violations and source identity survives multi-agent research handoffs.
B:In a vendor onboarding document pipeline, provenance metadata must remain attached when intermediate summaries are merged, but vague semantic similarity raises false positives; therefore the separate requirement that a classifier avoids marking uncertain cases as confirmed violations would remain unsatisfied.
C:In a vendor onboarding document pipeline, clear decision thresholds separate direct evidence from weak similarity, but the writer cannot reliably reconstruct which source supports each conclusion; therefore the separate requirement that source identity survives multi-agent research handoffs would remain unsatisfied.
Question 495
The architecture of a warranty claims assistant must meet these three criteria: a classifier avoids marking uncertain cases as confirmed violations; a failed retrieval agent does not masquerade as having found no matches; a failed extraction is retried with specific corrective information. Choose THREE measures, one for each requirement.
- define explicit positive evidence and an abstain or review category
- convert every retrieval failure into an empty list
- repeat the same vague prompt after each failure
- feed field-level validation errors back in a bounded retry
- return a typed failure status with its origin to the coordinator
- tell the model to flag anything that sounds suspicious
Explanation
Correct Answers
A:For a warranty claims assistant, clear decision thresholds separate direct evidence from weak similarity, which directly establishes that a classifier avoids marking uncertain cases as confirmed violations; the separate requirement that a failed retrieval agent does not masquerade as having found no matches is handled by a different action.
D:For a warranty claims assistant, targeted feedback identifies what the next attempt must correct, which directly establishes that a failed extraction is retried with specific corrective information; the separate requirement that a classifier avoids marking uncertain cases as confirmed violations is handled by a different action.
E:For a warranty claims assistant, an operational error must remain distinguishable from a valid empty result, which directly establishes that a failed retrieval agent does not masquerade as having found no matches; the separate requirement that a failed extraction is retried with specific corrective information is handled by a different action.
Incorrect Answers
B:For a warranty claims assistant, the final report could incorrectly claim no sources exist, so this option cannot establish that a failed retrieval agent does not masquerade as having found no matches; correctly addressing a failed extraction is retried with specific corrective information elsewhere does not repair that deficiency.
C:For a warranty claims assistant, unchanged requests do not reveal which constraint was violated, so this option cannot establish that a failed extraction is retried with specific corrective information; correctly addressing a classifier avoids marking uncertain cases as confirmed violations elsewhere does not repair that deficiency.
F:For a warranty claims assistant, vague semantic similarity raises false positives, so this option cannot establish that a classifier avoids marking uncertain cases as confirmed violations; correctly addressing a failed retrieval agent does not masquerade as having found no matches elsewhere does not repair that deficiency.
Question 496
Before expanding a pre-merge code inspection workflow, the owner must demonstrate that regressions introduced during refinement are detected and fixed, while also demonstrating that the implementation plan receives stakeholder sign-off. Which design provides both controls?
- restore or revise the problematic change and retest affected paths; and treat the agent's internally generated plan as implicit approval
- restore or revise the problematic change and retest affected paths; and present ordered changes, risks, and verification steps before switching to execution
- keep layering patches without checking the last stable behavior; and treat the agent's internally generated plan as implicit approval
- keep layering patches without checking the last stable behavior; and present ordered changes, risks, and verification steps before switching to execution
Explanation
Correct Answer
B:In a pre-merge code inspection workflow, iteration includes detecting regressions and returning to a passing state; alongside this, an explicit review point aligns the agent's work with the authorized scope, satisfying the distinct requirements that regressions introduced during refinement are detected and fixed and that the implementation plan receives stakeholder sign-off.
Incorrect Answers
A:In a pre-merge code inspection workflow, iteration includes detecting regressions and returning to a passing state, but the system cannot infer external authorization from its own reasoning; therefore the separate requirement that the implementation plan receives stakeholder sign-off would remain unsatisfied.
C:In a pre-merge code inspection workflow, cumulative changes can bury the source of failure and the system cannot infer external authorization from its own reasoning; the design would fail both conditions: regressions introduced during refinement are detected and fixed and the implementation plan receives stakeholder sign-off.
D:In a pre-merge code inspection workflow, an explicit review point aligns the agent's work with the authorized scope, 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.
Question 497
A team operating a contract-analysis research network wants to establish two separate behaviors: a delegated agent receives the background it needs to act; in addition, a straightforward lookup avoids unnecessary agent coordination. Which plan implements both?
- implement only pass the necessary task context and references explicitly at invocation; make no corresponding change for the second requirement
- assume the child agent sees all previous coordinator turns; and use one focused tool-assisted agent for the bounded task
- pass the necessary task context and references explicitly at invocation; and use one focused tool-assisted agent for the bounded task
- assume the child agent sees all previous coordinator turns; and spawn a coordinator plus multiple specialists for every trivial query
- pass the necessary task context and references explicitly at invocation; and spawn a coordinator plus multiple specialists for every trivial query
Explanation
Correct Answer
C:In a contract-analysis research network, subagents do not automatically inherit the coordinator's entire conversational history; alongside this, orchestration overhead is not justified when there is no useful specialization or concurrency, satisfying the distinct requirements that a delegated agent receives the background it needs to act and that a straightforward lookup avoids unnecessary agent coordination.
Incorrect Answers
A:In a contract-analysis research network, subagents do not automatically inherit the coordinator's entire conversational history, but the independent mechanism for the other criterion is absent: orchestration overhead is not justified when there is no useful specialization or concurrency; as written, the design does not ensure that a straightforward lookup avoids unnecessary agent coordination.
B:In a contract-analysis research network, orchestration overhead is not justified when there is no useful specialization or concurrency, but implicit context inheritance makes delegation fragile and incomplete; therefore the separate requirement that a delegated agent receives the background it needs to act would remain unsatisfied.
D:In a contract-analysis research network, implicit context inheritance makes delegation fragile and incomplete and extra agents add latency and handoff risk without new capabilities; the design would fail both conditions: a delegated agent receives the background it needs to act and a straightforward lookup avoids unnecessary agent coordination.
E:In a contract-analysis research network, subagents do not automatically inherit the coordinator's entire conversational history, but extra agents add latency and handoff risk without new capabilities; therefore the separate requirement that a straightforward lookup avoids unnecessary agent coordination would remain unsatisfied.
Question 498
For a change-impact analysis service, a reviewer identifies two independent obligations: a coding assistant first identifies relevant files in a large repository and tool responses are associated with the exact invocations that produced them. Which proposal should be approved?
- read the entire repository recursively into one prompt; and return tool results with newly invented identifiers
- discover paths using Glob before reading contents; and return tool results with newly invented identifiers
- discover paths using Glob before reading contents; and return tool_result blocks using the original tool_use_id values
- read the entire repository recursively into one prompt; and return tool_result blocks using the original tool_use_id values
Explanation
Correct Answer
C:In a change-impact analysis service, staged path discovery prevents unrelated modules from saturating the context; alongside this, correlation identifiers bind asynchronous or parallel results to the right requests, satisfying the distinct requirements that a coding assistant first identifies relevant files in a large repository and that tool responses are associated with the exact invocations that produced them.
Incorrect Answers
A:In a change-impact analysis service, large context dumps bury important code among irrelevant files and the model cannot reliably associate a result with an outstanding call; the design would fail both conditions: a coding assistant first identifies relevant files in a large repository and tool responses are associated with the exact invocations that produced them.
B:In a change-impact analysis service, staged path discovery prevents unrelated modules from saturating the context, 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.
D:In a change-impact analysis service, correlation identifiers bind asynchronous or parallel results to the right requests, but large context dumps bury important code among irrelevant files; therefore the separate requirement that a coding assistant first identifies relevant files in a large repository would remain unsatisfied.
Question 499
An incident postmortem for a policy document normalization service requires controls so that an uncertain high-impact decision is routed for review instead of automatic execution and that a classifier avoids marking uncertain cases as confirmed violations. Which paired remedy covers both failure modes?
- proceed whenever the assistant's prose sounds certain; and define explicit positive evidence and an abstain or review category
- set a clear evidence or confidence boundary and require review below it; and tell the model to flag anything that sounds suspicious
- set a clear evidence or confidence boundary and require review below it; and define explicit positive evidence and an abstain or review category
- proceed whenever the assistant's prose sounds certain; and tell the model to flag anything that sounds suspicious
Explanation
Correct Answer
C:In a policy document normalization service, risk-weighted gates limit autonomous actions when support is inadequate; alongside this, clear decision thresholds separate direct evidence from weak similarity, satisfying the distinct requirements that an uncertain high-impact decision is routed for review instead of automatic execution and that a classifier avoids marking uncertain cases as confirmed violations.
Incorrect Answers
A:In a policy document normalization service, clear decision thresholds separate direct evidence from weak similarity, 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.
B:In a policy document normalization service, risk-weighted gates limit autonomous actions when support is inadequate, but vague semantic similarity raises false positives; therefore the separate requirement that a classifier avoids marking uncertain cases as confirmed violations would remain unsatisfied.
D:In a policy document normalization service, fluent language is not a calibrated reliability measure and vague semantic similarity raises false positives; the design would fail both conditions: an uncertain high-impact decision is routed for review instead of automatic execution and a classifier avoids marking uncertain cases as confirmed violations.
Question 500
When designing a database migration coding project, the requirement is not satisfied unless two teams' analyses are reconciled before publication and a conversational task does not invoke tools needlessly. Which implementation addresses the two independent conditions?
- publish whichever branch finishes last; and force a tool on every message regardless of intent
- add a synthesis stage with explicit conflict-resolution criteria; and force a tool on every message regardless of intent
- add a synthesis stage with explicit conflict-resolution criteria; and use automatic tool choice when a call is optional
- publish whichever branch finishes last; and use automatic tool choice when a call is optional
Explanation
Correct Answer
C:In a database migration coding project, merging divergent work requires an authoritative step rather than choosing by arrival time; alongside this, auto allows the model to answer directly when no external side effect or fresh data is needed, satisfying the distinct requirements that two teams' analyses are reconciled before publication and that a conversational task does not invoke tools needlessly.
Incorrect Answers
A:In a database migration coding project, run order is not an evidence or quality standard and mandatory calls add latency and can invoke irrelevant operations; the design would fail both conditions: two teams' analyses are reconciled before publication and a conversational task does not invoke tools needlessly.
B:In a database migration coding project, merging divergent work requires an authoritative step rather than choosing by arrival time, but mandatory calls add latency and can invoke irrelevant operations; therefore the separate requirement that a conversational task does not invoke tools needlessly would remain unsatisfied.
D:In a database migration coding project, auto allows the model to answer directly when no external side effect or fresh data is needed, 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.