Microsoft AI-103 (Developing AI Apps and Agents on Azure) Practice Test 26

Question 250

The implementation does not meet its business requirement: The model invents API argument names and fails account-lookup requests repeatedly. Which corrective action is appropriate?

  1. Delimit external evidence and apply instruction-hierarchy and injection controls
  2. Specify the support role and allowed actions, and constrain available tools to that role
  3. Place a mandatory approval gate on the account-closure tool
  4. Provide a precise function description and schema with required parameters

Explanation

Correct Answer

D:

The fault signal is that the model invents API argument names and fails account-lookup requests repeatedly; the targeted fix works because a typed tool contract constrains argument names and shapes and simplifies validation; authorization still occurs on execution, restoring the ability to make the parameters of an account-lookup function explicit to the model and executor.

Incorrect Answers

A:

If the failing function were to keep untrusted customer-uploaded text from redefining the agent's permissions, this change could be relevant because trusted control instructions must have higher precedence than external content; retrieved pages are data, not agent policy; the actual symptom is that the model invents API argument names and fails account-lookup requests repeatedly, which calls for a different repair.

B:

If the failing function were to keep a support agent focused on diagnosing product issues rather than taking unrelated administrative action, this change could be relevant because a clear role and allowed-goal scope guide planning; enforceable permissions remain necessary at tool boundaries; the actual symptom is that the model invents API argument names and fails account-lookup requests repeatedly, which calls for a different repair.

C:

If the failing function were to prevent an autonomous agent from executing account closure without explicit review, this change could be relevant because a human approval checkpoint separates proposed actions from committed side effects and provides an auditable decision; the actual symptom is that the model invents API argument names and fails account-lookup requests repeatedly, which calls for a different repair.

Question 251

An engineering team has a requirement to resolve 'those two records' by retaining the preceding selection context. Which capability should be in the initial design?

  1. Expire temporary secrets and tool results
  2. Apply idempotency to write-capable tools
  3. Retrieve consented durable preferences when required
  4. Preserve short-term dialogue references

Explanation

Correct Answer

D:

For the stated requirement to resolve 'those two records' by retaining the preceding selection context, conversation tracking gives referential continuity across turns, but the context should be bounded to the correct user session; this is the mechanism that the system must implement.

Incorrect Answers

A:

For a system whose objective is to avoid retaining authentication tokens or temporary PII inside agent memory longer than needed, short-lived state and purpose-based retention protect secrets and sensitive data; summarization is not a substitute for removal; this does not directly meet the required outcome to resolve 'those two records' by retaining the preceding selection context.

B:

For a system whose objective is to prevent double charging when an agent retries after a network timeout, an idempotency key ties a logical transaction to one effect even if the same request is retried; this does not directly meet the required outcome to resolve 'those two records' by retaining the preceding selection context.

C:

For a system whose objective is to remember a customer's explicit notification preference across separate sessions, persistent preference storage is distinct from a single chat transcript and requires consent, access control, and deletion workflows; this does not directly meet the required outcome to resolve 'those two records' by retaining the preceding selection context.

Question 252

The implementation does not meet its business requirement: The agent answers from a stored shipping export even after the carrier updates its status. Which corrective action is appropriate?

  1. Route approved computation to an isolated code tool with validated file inputs
  2. Register the retrieval capability in the agent workflow and attach returned source passages before response generation
  3. Use stable operation identifiers and enforce idempotent processing server-side
  4. Use a permitted status lookup tool with the customer's authorization context

Explanation

Correct Answer

D:

The fault signal is that the agent answers from a stored shipping export even after the carrier updates its status; the targeted fix works because a scoped API tool obtains current transactional facts while maintaining least-privilege read access, restoring the ability to let an agent check a shipment's present location without permission to modify the order.

Incorrect Answers

A:

If the failing function were to let an agent aggregate a supplied numeric CSV where exact computation is important, this change could be relevant because a sandboxed execution tool performs calculations deterministically; executable code needs strict isolation and resource limits; the actual symptom is that the agent answers from a stored shipping export even after the carrier updates its status, which calls for a different repair.

B:

If the failing function were to invoke governed document search as a tool at the point the agent needs evidence for an answer, this change could be relevant because an agent run needs explicit retrieval calls and returned context before composing a grounded response; model parameters alone do not fetch current documentation; the actual symptom is that the agent answers from a stored shipping export even after the carrier updates its status, which calls for a different repair.

C:

If the failing function were to prevent double charging when an agent retries after a network timeout, this change could be relevant because an idempotency key ties a logical transaction to one effect even if the same request is retried; the actual symptom is that the agent answers from a stored shipping export even after the carrier updates its status, which calls for a different repair.

Question 253

The acceptance criteria for a Foundry-based application require it to invoke governed document search as a tool at the point the agent needs evidence for an answer. Which design should be approved?

  1. Register a read-only API tool for live status
  2. Limit information passed between specialist agents
  3. Trace step-level agent decisions and failures
  4. Use a content-extraction tool for uploaded forms
  5. Orchestrate agent knowledge lookup during a conversation

Explanation

Correct Answer

E:

For the stated requirement to invoke governed document search as a tool at the point the agent needs evidence for an answer, an agent run needs explicit retrieval calls and returned context before composing a grounded response; model parameters alone do not fetch current documentation; this is the mechanism that the system must implement.

Incorrect Answers

A:

For a system whose objective is to let an agent check a shipment's present location without permission to modify the order, a scoped API tool obtains current transactional facts while maintaining least-privilege read access; this does not directly meet the required outcome to invoke governed document search as a tool at the point the agent needs evidence for an answer.

B:

For a system whose objective is to stop one agent from receiving private data that another agent needed only for its subtask, scoped messages and redacted handoff contracts minimize sensitive context propagation across multiple agents; this does not directly meet the required outcome to invoke governed document search as a tool at the point the agent needs evidence for an answer.

C:

For a system whose objective is to find where a lengthy agent plan diverges from intended workflow, per-step tracing captures model requests, tool parameters, statuses and transitions; a final response log hides the path taken; this does not directly meet the required outcome to invoke governed document search as a tool at the point the agent needs evidence for an answer.

D:

For a system whose objective is to enable an agent to read key fields from photographed or scanned paperwork, document extraction tools convert visual text and layout into structured data suitable for subsequent agent reasoning; this does not directly meet the required outcome to invoke governed document search as a tool at the point the agent needs evidence for an answer.

Question 254

Production telemetry reveals this failure: A returning user has to repeat an explicitly stored setting after each new session. Which change most directly corrects it?

  1. Load only authorized durable preferences into the current session context
  2. Exclude transient secrets from memory and enforce state expiration policies
  3. Carry relevant prior tool results and user selections in session-scoped conversation state
  4. Place a mandatory approval gate on the account-closure tool

Explanation

Correct Answer

A:

The fault signal is that a returning user has to repeat an explicitly stored setting after each new session; the targeted fix works because persistent preference storage is distinct from a single chat transcript and requires consent, access control, and deletion workflows, restoring the ability to remember a customer's explicit notification preference across separate sessions.

Incorrect Answers

B:

If the failing function were to avoid retaining authentication tokens or temporary PII inside agent memory longer than needed, this change could be relevant because short-lived state and purpose-based retention protect secrets and sensitive data; summarization is not a substitute for removal; the actual symptom is that a returning user has to repeat an explicitly stored setting after each new session, which calls for a different repair.

C:

If the failing function were to resolve 'those two records' by retaining the preceding selection context, this change could be relevant because conversation tracking gives referential continuity across turns, but the context should be bounded to the correct user session; the actual symptom is that a returning user has to repeat an explicitly stored setting after each new session, which calls for a different repair.

D:

If the failing function were to prevent an autonomous agent from executing account closure without explicit review, this change could be relevant because a human approval checkpoint separates proposed actions from committed side effects and provides an auditable decision; the actual symptom is that a returning user has to repeat an explicitly stored setting after each new session, which calls for a different repair.

Question 255

Engineers have implemented a change to define an agent's bounded role and objective. What validation would expose a remaining defect?

  1. Replay interrupted workflows and verify each external side effect occurs once
  2. Inspect inter-agent messages for unauthorized fields and verify output correctness
  3. Test out-of-scope requests for proper refusal or delegated escalation
  4. Red-team embedded document instructions and inspect unauthorized tool-call attempts

Explanation

Correct Answer

C:

The accepted change is to define an agent's bounded role and objective; this evaluation checks whether it can keep a support agent focused on diagnosing product issues rather than taking unrelated administrative action, since a clear role and allowed-goal scope guide planning; enforceable permissions remain necessary at tool boundaries.

Incorrect Answers

A:

This test concerns the ability to resume a long-running workflow safely after an inference retry, for which durable workflow state records completed steps and pending approvals; relying only on the model's last text risks duplicate side effects; it does not establish whether the change to define an agent's bounded role and objective works for the current release.

B:

This test concerns the ability to stop one agent from receiving private data that another agent needed only for its subtask, for which scoped messages and redacted handoff contracts minimize sensitive context propagation across multiple agents; it does not establish whether the change to define an agent's bounded role and objective works for the current release.

D:

This test concerns the ability to keep untrusted customer-uploaded text from redefining the agent's permissions, for which trusted control instructions must have higher precedence than external content; retrieved pages are data, not agent policy; it does not establish whether the change to define an agent's bounded role and objective works for the current release.

Question 256

After implementing a workflow to decommission obsolete agent versions safely, the team needs a robust release gate. Which check should it choose?

  1. Use a labeled error taxonomy and compare the distribution before and after each change
  2. Monitor calls to the retired version and test that the old tool endpoint is inaccessible
  3. Attempt cross-user requests and confirm access denials in both retrieval and tools
  4. Audit per-agent tool entitlements and cross-service invocation traces

Explanation

Correct Answer

B:

The accepted change is to decommission obsolete agent versions safely; this evaluation checks whether it can remove old versions that retain deprecated tools without disrupting active clients, since a controlled retirement checks clients, revokes unused credentials, and archives relevant audit records before deletion.

Incorrect Answers

A:

This test concerns the ability to prioritize corrections across retrieval misses, malformed tools, policy blocks, and model planning errors, for which failure classification links observed symptoms to the component that can be improved; a single aggregate error rate obscures causes; it does not establish whether the change to decommission obsolete agent versions safely works for the current release.

C:

This test concerns the ability to enforce what both the service and the delegated user are authorized to do, for which a workload identity authenticates the agent service, while user-aware authorization limits operations on behalf of a caller; it does not establish whether the change to decommission obsolete agent versions safely works for the current release.

D:

This test concerns the ability to avoid giving every agent separate uncontrolled credentials for the same external service, for which centralized governed tool access can standardize authentication and logging while each agent remains least privileged; it does not establish whether the change to decommission obsolete agent versions safely works for the current release.

Question 257

After deployment, the team observes: A payment tool times out after charging, and the agent retries the charge as if no payment happened. Which remediation addresses the likely cause?

  1. Invoke an authorized OCR-and-layout extraction tool before applying business rules
  2. Define and test compensating operations for successfully completed steps when later steps fail
  3. Use stable operation identifiers and enforce idempotent processing server-side
  4. Place a mandatory approval gate on the account-closure tool

Explanation

Correct Answer

C:

The fault signal is that a payment tool times out after charging, and the agent retries the charge as if no payment happened; the targeted fix works because an idempotency key ties a logical transaction to one effect even if the same request is retried, restoring the ability to prevent double charging when an agent retries after a network timeout.

Incorrect Answers

A:

If the failing function were to enable an agent to read key fields from photographed or scanned paperwork, this change could be relevant because document extraction tools convert visual text and layout into structured data suitable for subsequent agent reasoning; the actual symptom is that a payment tool times out after charging, and the agent retries the charge as if no payment happened, which calls for a different repair.

B:

If the failing function were to recover from a partially completed workflow when a later operation fails, this change could be relevant because a compensating action undoes or offsets previously committed steps where a transaction cannot be atomic across services; the actual symptom is that a payment tool times out after charging, and the agent retries the charge as if no payment happened, which calls for a different repair.

D:

If the failing function were to prevent an autonomous agent from executing account closure without explicit review, this change could be relevant because a human approval checkpoint separates proposed actions from committed side effects and provides an auditable decision; the actual symptom is that a payment tool times out after charging, and the agent retries the charge as if no payment happened, which calls for a different repair.

Question 258

An Azure AI workload passes basic health probes but has this issue: The model summarizes a table but makes arithmetic errors in multi-column totals. Which fix targets it?

  1. Route approved computation to an isolated code tool with validated file inputs
  2. Place a mandatory approval gate on the account-closure tool
  3. Invoke an authorized OCR-and-layout extraction tool before applying business rules
  4. Register the retrieval capability in the agent workflow and attach returned source passages before response generation

Explanation

Correct Answer

A:

The fault signal is that the model summarizes a table but makes arithmetic errors in multi-column totals; the targeted fix works because a sandboxed execution tool performs calculations deterministically; executable code needs strict isolation and resource limits, restoring the ability to let an agent aggregate a supplied numeric CSV where exact computation is important.

Incorrect Answers

B:

If the failing function were to prevent an autonomous agent from executing account closure without explicit review, this change could be relevant because a human approval checkpoint separates proposed actions from committed side effects and provides an auditable decision; the actual symptom is that the model summarizes a table but makes arithmetic errors in multi-column totals, which calls for a different repair.

C:

If the failing function were to enable an agent to read key fields from photographed or scanned paperwork, this change could be relevant because document extraction tools convert visual text and layout into structured data suitable for subsequent agent reasoning; the actual symptom is that the model summarizes a table but makes arithmetic errors in multi-column totals, which calls for a different repair.

D:

If the failing function were to invoke governed document search as a tool at the point the agent needs evidence for an answer, this change could be relevant because an agent run needs explicit retrieval calls and returned context before composing a grounded response; model parameters alone do not fetch current documentation; the actual symptom is that the model summarizes a table but makes arithmetic errors in multi-column totals, which calls for a different repair.

Question 259

After deployment, the team observes: An internal agent uses its broad service permissions to return another user's private account data. Which remediation addresses the likely cause?

  1. Use an approved shared tool integration with centralized access policy and secret handling
  2. Enable correlated tracing across decision and tool spans with appropriate redaction
  3. Migrate consumers, disable the stale agent endpoint, and revoke its credentials
  4. Propagate the caller's security context and authorize each sensitive query appropriately

Explanation

Correct Answer

D:

The fault signal is that an internal agent uses its broad service permissions to return another user's private account data; the targeted fix works because a workload identity authenticates the agent service, while user-aware authorization limits operations on behalf of a caller, restoring the ability to enforce what both the service and the delegated user are authorized to do.

Incorrect Answers

A:

If the failing function were to avoid giving every agent separate uncontrolled credentials for the same external service, this change could be relevant because centralized governed tool access can standardize authentication and logging while each agent remains least privileged; the actual symptom is that an internal agent uses its broad service permissions to return another user's private account data, which calls for a different repair.

B:

If the failing function were to find where a lengthy agent plan diverges from intended workflow, this change could be relevant because per-step tracing captures model requests, tool parameters, statuses and transitions; a final response log hides the path taken; the actual symptom is that an internal agent uses its broad service permissions to return another user's private account data, which calls for a different repair.

C:

If the failing function were to remove old versions that retain deprecated tools without disrupting active clients, this change could be relevant because a controlled retirement checks clients, revokes unused credentials, and archives relevant audit records before deletion; the actual symptom is that an internal agent uses its broad service permissions to return another user's private account data, which calls for a different repair.

Question 260

A user-acceptance test fails with the following observation: The writing agent receives raw salary records even though it needs only aggregate statistics. What is the best correction?

  1. Add a supervisor that gathers evidence, resolves disagreements, and approves final synthesis
  2. Pass sanitized aggregate outputs and enforce each agent's data entitlements
  3. Delimit external evidence and apply instruction-hierarchy and injection controls
  4. Classify request intent and dispatch to an authorized specialist with scoped capabilities

Explanation

Correct Answer

B:

The fault signal is that the writing agent receives raw salary records even though it needs only aggregate statistics; the targeted fix works because scoped messages and redacted handoff contracts minimize sensitive context propagation across multiple agents, restoring the ability to stop one agent from receiving private data that another agent needed only for its subtask.

Incorrect Answers

A:

If the failing function were to combine independent research subtasks into a coherent final response with oversight, this change could be relevant because a supervisor can assign independent tasks, collect outputs, and apply cross-checks before presenting an answer; the actual symptom is that the writing agent receives raw salary records even though it needs only aggregate statistics, which calls for a different repair.

C:

If the failing function were to keep untrusted customer-uploaded text from redefining the agent's permissions, this change could be relevant because trusted control instructions must have higher precedence than external content; retrieved pages are data, not agent policy; the actual symptom is that the writing agent receives raw salary records even though it needs only aggregate statistics, which calls for a different repair.

D:

If the failing function were to route finance, operations, and security questions to different bounded expert agents, this change could be relevant because a router delegates to an appropriate specialist with a narrow tool set; a single all-powerful agent is harder to govern; the actual symptom is that the writing agent receives raw salary records even though it needs only aggregate statistics, which calls for a different repair.

Developing AI Apps and Agents on Azure (AI-103) Practice Tests