Question 130
A design adds autonomous behavior to a process where users use novel requests outside the supported topic inventory. This creates a documented risk: An unmatched fallback might accidentally bypass required identity verification. Which enforced control is the best fit?
- Honor source access when syncing or federating content
- Validate types and authorization at the called service
- Stop and compensate when a required approval fails
- Prevent fallback from bypassing authentication or policies
Explanation
Correct Answer
D:In the context where users use novel requests outside the supported topic inventory, the exposure is that an unmatched fallback might accidentally bypass required identity verification; containing it requires a guardrail in the operation where fallback handles unmatched intent and can ask for clarification or escalate.
Incorrect Answers
A:This restriction is associated with workers need internal wiki articles from outside Microsoft 365 and addresses the exposure that an external connector might index documents without honoring original ACLs; it does not specifically protect the target workflow against an unmatched fallback might accidentally bypass required identity verification.
B:Another process could need this guardrail because an API might accept a valid-looking but unauthorized generated identifier; the design being assessed supports users use novel requests outside the supported topic inventory and has the explicit exposure that an unmatched fallback might accidentally bypass required identity verification.
C:Where approval processing requires sequential verification and notification, the guard can address a flow might notify the customer even though a required approval failed; the exposure that an unmatched fallback might accidentally bypass required identity verification remains relevant to the present assessment.
Question 131
A staging test has to detect the specific defect that the agent fabricates an answer during a search outage. In the workflow where required knowledge stores are intermittently unreachable, which exercise verifies the corrective behavior?
- Ask about a revised policy and verify an allowed citation
- Change a status and verify immediate answer consistency
- Disconnect the source and verify a transparent failure response
- Inject hostile instructions into a test document and check refusal
Explanation
Correct Answer
C:For the required workflow where required knowledge stores are intermittently unreachable, the test needs to reproduce or prevent the failure where the agent fabricates an answer during a search outage; disconnecting the source and verifying a transparent failure response exercises the path where health checks and explicit no-evidence responses handle missing context safely.
Incorrect Answers
A:The suggested test serves employee policies reside in governed Microsoft 365 documents, in which answers omit current employee handbook changes; the present scenario instead concerns the reported problem where the agent fabricates an answer during a search outage.
B:Although the exercise might catch a cached knowledge answer is several hours old, it would not directly validate the mechanism that health checks and explicit no-evidence responses handle missing context safely in the target implementation.
D:This check exercises a different system behavior: trust boundaries treat source text as evidence rather than authority; a pass there cannot establish that the agent fabricates an answer during a search outage no longer occurs in the required workflow.
Question 132
A business stakeholder describes this operating situation: An autonomous agent relies on slow approvals and delayed tool results. Testing instead shows that timeout creates an abandoned workflow with no owner. What design approach is most appropriate?
- Track successful task completion rather than conversational volume
- Test multi-turn context retention and clarification
- Test long-running and partially completed plans
- Test resilience when a grounding dependency fails
Explanation
Correct Answer
C:The mechanism that state-machine tests examine retries, expiry and compensation fits the scenario where an autonomous agent relies on slow approvals and delayed tool results; it addresses the defect that timeout creates an abandoned workflow with no owner.
Incorrect Answers
A:For another type of architecture, outcome measures connect agent use with completed business work is useful; this design must instead support an autonomous agent relies on slow approvals and delayed tool results despite the problem where timeout creates an abandoned workflow with no owner.
B:In the case where agent must remember order context while asking for missing details, the alternate design depends on multi-turn tests assess state tracking, entity separation and follow-up questions; it does not correct the target failure, timeout creates an abandoned workflow with no owner.
D:The agent invents answers rather than saying evidence is missing; the alternate design can address that situation because failure injection checks controlled degradation and abstention, whereas this question needs a solution to timeout creates an abandoned workflow with no owner.
Question 133
Troubleshooting starts from a business complaint: The final order lacks traceability to the AI suggestion and approver. The affected process is: Service recommends a price adjustment requiring finance approval. Which investigation is most directly tied to the reported defect?
- Audit send events and approver identities after rollout
- Compare suggested orders and actual planner decisions by cohort
- Reconcile audit IDs across each application
- Compare generated tests to authentication and schema contracts
Explanation
Correct Answer
C:The investigation directly tests the causal connection: transaction-path tests confirm recommendation, review and posting records; it collects evidence for a workflow where service recommends a price adjustment requiring finance approval but the outcome is instead that the final order lacks traceability to the AI suggestion and approver.
Incorrect Answers
A:New prompt automation bypasses supervisor approval; examining that other path does not determine whether transaction-path tests confirm recommendation, review and posting records, which is what distinguishes the issue in the stated use case.
B:An AI-driven replenishment proposal influences procurement; its matching diagnostic check is reasonable for recommendation changes spread to all planners without evaluation; the present investigation needs evidence for the different failure where the final order lacks traceability to the AI suggestion and approver.
D:Agents can call APIs with customer IDs and amount values; its matching diagnostic check is reasonable for happy-path fixtures omit unauthorized and invalid inputs; the present investigation needs evidence for the different failure where the final order lacks traceability to the AI suggestion and approver.
Question 134
The operating team must monitor whether an architectural correction works. Business context: Agents suggest payment adjustments and inventory holds. Historical symptom: A new connector allows a service identity to post directly. Which outcome measure is the best match?
- Mapped operational fields reconciled across releases
- Cross-app exceptions ending in consistent state
- Unauthorized finance postings in release tests
- Model deployments passing artifact-integrity controls
Explanation
Correct Answer
C:For the business use where agents suggest payment adjustments and inventory holds, the result to quantify is unauthorized finance postings in release tests; it connects measurement to the mechanism where authorization regression gates verify posting authority remains constrained, not merely to agent traffic.
Incorrect Answers
A:Although mapped operational fields reconciled across releases is observable, the affected process involves agents suggest payment adjustments and inventory holds; its acceptance outcome is unauthorized finance postings in release tests, not the alternative measurement.
B:That scorecard concerns the activity in which a blocked invoice triggers a hold on related procurement actions and its result of cross-app exceptions ending in consistent state; the reported shortcoming here is a new connector allows a service identity to post directly.
D:A change in model deployments passing artifact-integrity controls is informative for deployment automation pulls model artifacts from unverified origins; it does not establish whether the current process achieves unauthorized finance postings in release tests.
Question 135
An incident review has one confirmed symptom: Combined metrics hide voice users trapped in fallback loops. For the use case where chat and voice agents have different interaction patterns, which targeted diagnostic check would reveal the most pertinent evidence?
- Compare errors by relevant justified cohorts with privacy care
- Compare abandonment and handoff reasons by channel
- Compare outcomes by locale, channel and agent version
- Compare operation keys with final business transaction counts
- Verify spans and errors arrive in the expected telemetry resource
Explanation
Correct Answer
B:Where combined metrics hide voice users trapped in fallback loops, the useful evidence comes from comparing abandonment and handoff reasons by channel; its focus matches the architectural mechanism that channel-specific analytics reveal where interactions fail in the process where chat and voice agents have different interaction patterns.
Incorrect Answers
A:In the separate scenario where a classifier underperforms on one language group, the same investigation could explain overall accuracy conceals disproportionate error; it does not isolate the reported cause in a system where chat and voice agents have different interaction patterns.
C:In the separate scenario where a conversational change affects only one support geography, the same investigation could explain global averages hide a regional escalation spike; it does not isolate the reported cause in a system where chat and voice agents have different interaction patterns.
D:For the failure that parallel agents each submit a costly replacement, comparing operation keys with final business transaction counts is relevant; this question instead reports that combined metrics hide voice users trapped in fallback loops, so different evidence is needed.
E:Verifying spans and errors arrive in the expected telemetry resource explores a separate connection, where connected observability supplies traces and operational metrics; the diagnosed requirement is chat and voice agents have different interaction patterns, and the actual defect is combined metrics hide voice users trapped in fallback loops.
Question 136
The architecture board has evidence that transcription confusion risks wrong account modifications. The affected situation is: Callers provide account numbers and amounts by speech. What change most directly establishes the missing behavior?
- Use MCP tools for a governed reusable integration surface
- Design voice interaction with confirmation of critical entities
- Set autonomy boundaries with approval checkpoints
- Use a governed connector for live external account data
Explanation
Correct Answer
B:For the need that callers provide account numbers and amounts by speech, the relevant mechanism is that spoken confirmation and verification reduce high-impact recognition errors; that architectural dependency addresses the observed failure where transcription confusion risks wrong account modifications.
Incorrect Answers
A:For another type of architecture, a supported MCP server exposes typed tool capabilities to clients is useful; this design must instead support callers provide account numbers and amounts by speech despite the problem where transcription confusion risks wrong account modifications.
C:This design would make sense if agent can research options but cannot approve a vendor payment, particularly where a background process posts expenses beyond its delegated scope; the requested architecture instead needs the dependency that spoken confirmation and verification reduce high-impact recognition errors because transcription confusion risks wrong account modifications.
D:When sellers need current credit exposure held outside CRM, a supported connector can retrieve external information with explicit identity mapping; in the stated workflow, however, transcription confusion risks wrong account modifications; the relevant architectural behavior is that spoken confirmation and verification reduce high-impact recognition errors.
Question 137
Troubleshooting starts from a business complaint: A Copilot suggestion includes restricted invoice details. The affected process is: Sales and Finance roles have different record entitlements. Which investigation is most directly tied to the reported defect?
- Review backend authorization logs against actor identities
- Review virtual-entity metadata and source-field changes
- Compare deployment logs with financial close freezes
- Compare retrieval and tool responses using two user roles
Explanation
Correct Answer
D:For the situation where sales and Finance roles have different record entitlements, the defect is that a Copilot suggestion includes restricted invoice details; comparing retrieval and tool responses using two user roles helps verify the cause-and-effect path: cross-role tests verify that authorization survives integration.
Incorrect Answers
A:That probe evaluates server-side access checks enforce real actor entitlements in a workflow where prompt wording persuades the agent to submit a privileged operation; the reported failure concerns the path for which cross-role tests verify that authorization survives integration instead.
B:For the failure that an update changes a field mapping and grounded answers become stale, reviewing virtual-entity metadata and source-field changes is relevant; this question instead reports that a Copilot suggestion includes restricted invoice details, so different evidence is needed.
C:That probe evaluates coordinated release windows account for transaction criticality in a workflow where invoice posting agents affect critical close processes; the reported failure concerns the path for which cross-role tests verify that authorization survives integration instead.
Question 138
Troubleshooting starts from a business complaint: A prompt change during close generates unexplained exceptions. The affected process is: Invoice posting agents affect critical close processes. Which investigation is most directly tied to the reported defect?
- Compare deployment logs with financial close freezes
- Review endpoint access logs and allowed network paths
- Compare ticket assignment confusion matrices across versions
- Inspect solution dependencies and common entity schema
Explanation
Correct Answer
A:For the case where invoice posting agents affect critical close processes, the diagnostic path must explain the failure that a prompt change during close generates unexplained exceptions; comparing deployment logs with financial close freezes targets the mechanism by which coordinated release windows account for transaction criticality.
Incorrect Answers
B:The evidence this method produces applies to a workflow where a sensitive model endpoint is reachable using a broadly shared key; it cannot by itself establish why a prompt change during close generates unexplained exceptions in the current scenario.
C:In the process where a support agent changes how it classifies priority intents, the diagnostic procedure could explain high-severity tickets begin routing to general queues; this particular incident instead turns on whether coordinated release windows account for transaction criticality and why a prompt change during close generates unexplained exceptions.
D:Inspecting solution dependencies and common entity schema explores a separate connection, where cross-app release plans validate shared contracts and timing; the diagnosed requirement is invoice posting agents affect critical close processes, and the actual defect is a prompt change during close generates unexplained exceptions.
Question 139
After deployment, stakeholders need an interpretable measure for this use case: Fixes are marked complete when a prompt is deployed. The pilot showed that previously reported errors recur in the next release. Which metric captures the needed result?
- Unapproved external destinations reached through agent tools
- Median time to correctly triage an agent incident
- Percentage of fixes that pass production effectiveness checks
- Authorized tool calls succeeding after promotion
Explanation
Correct Answer
C:Because outcome verification checks whether the change fixed the user problem, the outcome worth following is percentage of fixes that pass production effectiveness checks; the metric is tied to the reported issue—previously reported errors recur in the next release—for the scenario where fixes are marked complete when a prompt is deployed.
Incorrect Answers
A:For the situation where a prompt injection requests data export to an outside address, unapproved external destinations reached through agent tools is a sensible indicator; the problem that previously reported errors recur in the next release requires measurement of percentage of fixes that pass production effectiveness checks.
B:For the workflow where complaints use vague labels such as agent is wrong, median time to correctly triage an agent incident is an actionable measure; it does not establish whether previously reported errors recur in the next release in the situation under review.
D:Although authorized tool calls succeeding after promotion is observable, the affected process involves fixes are marked complete when a prompt is deployed; its acceptance outcome is percentage of fixes that pass production effectiveness checks, not the alternative measurement.
Question 140
An incident review has one confirmed symptom: Small-model outputs omit critical contractual exceptions. For the use case where some cases need cross-document reasoning and detailed justification, which targeted diagnostic check would reveal the most pertinent evidence?
- Review escalated case outcomes and failure patterns
- Inspect join keys and collision rates in the ingestion pipeline
- Audit whether failures concern style rather than missing current facts
- Compare host limitations against the requirements matrix
Explanation
Correct Answer
A:The reported fault is that small-model outputs omit critical contractual exceptions in a system where some cases need cross-document reasoning and detailed justification; evidence from reviewing escalated case outcomes and failure patterns can show whether complexity classification selects a capable model for difficult tasks.
Incorrect Answers
B:That probe evaluates canonical keys and deduplication prevent incorrect cross-record joins in a workflow where multiple systems use inconsistent customer identifiers and duplicate cases; the reported failure concerns the path for which complexity classification selects a capable model for difficult tasks instead.
C:The proposed probe concerns a stable specialized classification style persists despite prompt examples, where fine-tuning can adapt recurring behavior using vetted examples; this incident instead requires evidence about complexity classification selects a capable model for difficult tasks and the failure that small-model outputs omit critical contractual exceptions.
D:That probe evaluates owning the engine provides control over models, routing and execution in a workflow where a workflow needs proprietary scheduling, model choice and runtime control; the reported failure concerns the path for which complexity classification selects a capable model for difficult tasks instead.