Question 70
An architecture review of a production API product lists two distinct outcomes: the first subsystem must extract claims from a document while retaining its visual layout, and a second subsystem must distinguish an invalid request from a temporary provider or rate-limit failure. Which pair delivers these outcomes?
- First component: Supply a tool definition with input schema. Second component: Read the HTTP error status before retrying.
- First component: Include an image content block. Second component: Supply a tool definition with input schema.
- First component: Use supported document or PDF input. Second component: Include an image content block.
- First component: Use supported document or PDF input. Second component: Read the HTTP error status before retrying.
Explanation
Correct Answer
D:The first proposal fits its assigned role because document-aware input lets a capable model consider page content and graphics when supported, while the second proposal fits its assigned role because status codes and structured errors guide correction versus bounded retry; their respective acceptance criteria are to extract claims from a document while retaining its visual layout and to distinguish an invalid request from a temporary provider or rate-limit failure.
Incorrect Answers
A:The first proposal serves the different purpose to let Claude request a read-only inventory lookup with typed arguments instead of the assigned first role because a declared tool name, description, and schema constrain the interface offered to the model, while the second proposal fits its assigned role because status codes and structured errors guide correction versus bounded retry; their respective acceptance criteria are to extract claims from a document while retaining its visual layout and to distinguish an invalid request from a temporary provider or rate-limit failure.
B:The first proposal serves the different purpose to analyze a photographed shipping label with the supported multimodal Messages input instead of the assigned first role because image inputs are supplied as typed content with appropriate source data or references, while the second proposal serves the different purpose to let Claude request a read-only inventory lookup with typed arguments instead of the assigned second role because a declared tool name, description, and schema constrain the interface offered to the model; their respective acceptance criteria are to extract claims from a document while retaining its visual layout and to distinguish an invalid request from a temporary provider or rate-limit failure.
C:The first proposal fits its assigned role because document-aware input lets a capable model consider page content and graphics when supported, while the second proposal serves the different purpose to analyze a photographed shipping label with the supported multimodal Messages input instead of the assigned second role because image inputs are supplied as typed content with appropriate source data or references; their respective acceptance criteria are to extract claims from a document while retaining its visual layout and to distinguish an invalid request from a temporary provider or rate-limit failure.
Question 71
A design review identifies a missing capability in a language-model integration project: it must distinguish borderline policy labels with examples of both allowed and disallowed cases. Which change meets that goal?
- Recognize next-token generation
- Use one representative example
- Distinguish input from output token limits
- Use contrasting few-shot examples
Explanation
Correct Answer
D:Multiple representative cases help define a decision boundary and reduce ambiguity; this behavior directly supports the need to distinguish borderline policy labels with examples of both allowed and disallowed cases. Repeating near-identical samples does not cover edge cases.
Incorrect Answers
A:When an application needs to explain why a plausible answer can still contain an unsupported factual claim, the model predicts likely continuations conditioned on context rather than querying an authoritative truth database by default; the current capability must instead distinguish borderline policy labels with examples of both allowed and disallowed cases, while fluent prose is not evidence that a source was verified.
B:A single demonstration can anchor the expected shape and tone of a generated response; the supported outcome is to clarify a specific formatting convention that a plain instruction was not conveying, whereas this scenario requires the ability to distinguish borderline policy labels with examples of both allowed and disallowed cases; an example is not a formal guarantee that every response will conform.
C:This approach serves the separate need to avoid confusing allowable prompt size with maximum generated response length because the context budget constrains total material considered while output ceilings bound what a single generation can emit; it does not establish the outcome to distinguish borderline policy labels with examples of both allowed and disallowed cases, and raising an output cap does not make an oversized input fit.
Question 72
Two features of a retrieval pipeline system are under review. One feature must keep an embedded webpage instruction from becoming higher-priority application policy; the other must avoid answering a pricing question from last month's retrieved cache. Which plan matches the respective needs?
- First component: Preserve instruction and evidence boundaries. Second component: Prune obsolete tool outputs.
- First component: Prune obsolete tool outputs. Second component: Anchor summaries to source identifiers.
- First component: Preserve instruction and evidence boundaries. Second component: Refresh stale evidence before reuse.
- First component: Anchor summaries to source identifiers. Second component: Refresh stale evidence before reuse.
Explanation
Correct Answer
C:The first proposal fits its assigned role because context assembly must label external text as data and avoid granting it authority over tool use, while the second proposal fits its assigned role because freshness checks determine whether a saved passage remains authoritative for a dynamic fact; their respective acceptance criteria are to keep an embedded webpage instruction from becoming higher-priority application policy and to avoid answering a pricing question from last month's retrieved cache.
Incorrect Answers
A:The first proposal fits its assigned role because context assembly must label external text as data and avoid granting it authority over tool use, while the second proposal serves the different purpose to stop stale log dumps from crowding out new evidence during an agent session instead of the assigned second role because tool-result pruning removes redundant bulky observations after their findings are retained; their respective acceptance criteria are to keep an embedded webpage instruction from becoming higher-priority application policy and to avoid answering a pricing question from last month's retrieved cache.
B:The first proposal serves the different purpose to stop stale log dumps from crowding out new evidence during an agent session instead of the assigned first role because tool-result pruning removes redundant bulky observations after their findings are retained, while the second proposal serves the different purpose to enable a reviewer to reconstruct evidence after context has been compacted instead of the assigned second role because stable document IDs and passage references preserve provenance beyond the shortened working memory; their respective acceptance criteria are to keep an embedded webpage instruction from becoming higher-priority application policy and to avoid answering a pricing question from last month's retrieved cache.
D:The first proposal serves the different purpose to enable a reviewer to reconstruct evidence after context has been compacted instead of the assigned first role because stable document IDs and passage references preserve provenance beyond the shortened working memory, while the second proposal fits its assigned role because freshness checks determine whether a saved passage remains authoritative for a dynamic fact; their respective acceptance criteria are to keep an embedded webpage instruction from becoming higher-priority application policy and to avoid answering a pricing question from last month's retrieved cache.
Question 73
A Claude application team owns two independent components. The first must let Claude request a read-only inventory lookup with typed arguments; the second must actually retrieve a customer's order status requested by Claude. Which paired plan fulfills both?
- First component: Read the HTTP error status before retrying. Second component: Process tool_use responses in the application.
- First component: Handle the tool_use stop_reason. Second component: Read the HTTP error status before retrying.
- First component: Supply a tool definition with input schema. Second component: Process tool_use responses in the application.
- First component: Supply a tool definition with input schema. Second component: Handle the tool_use stop_reason.
Explanation
Correct Answer
C:The first proposal fits its assigned role because a declared tool name, description, and schema constrain the interface offered to the model, while the second proposal fits its assigned role because client tools execute in application code; the model emits structured requests but does not invoke external functions itself; their respective acceptance criteria are to let Claude request a read-only inventory lookup with typed arguments and to actually retrieve a customer's order status requested by Claude.
Incorrect Answers
A:The first proposal serves the different purpose to distinguish an invalid request from a temporary provider or rate-limit failure instead of the assigned first role because status codes and structured errors guide correction versus bounded retry, while the second proposal fits its assigned role because client tools execute in application code; the model emits structured requests but does not invoke external functions itself; their respective acceptance criteria are to let Claude request a read-only inventory lookup with typed arguments and to actually retrieve a customer's order status requested by Claude.
B:The first proposal serves the different purpose to continue the agent loop when Claude is waiting for external observations instead of the assigned first role because this stop reason means tool calls need to be executed and their results returned before synthesis, while the second proposal serves the different purpose to distinguish an invalid request from a temporary provider or rate-limit failure instead of the assigned second role because status codes and structured errors guide correction versus bounded retry; their respective acceptance criteria are to let Claude request a read-only inventory lookup with typed arguments and to actually retrieve a customer's order status requested by Claude.
D:The first proposal fits its assigned role because a declared tool name, description, and schema constrain the interface offered to the model, while the second proposal serves the different purpose to continue the agent loop when Claude is waiting for external observations instead of the assigned second role because this stop reason means tool calls need to be executed and their results returned before synthesis; their respective acceptance criteria are to let Claude request a read-only inventory lookup with typed arguments and to actually retrieve a customer's order status requested by Claude.
Question 74
During acceptance testing for an agent governance service, one requirement is to ensure a policy block is interpreted correctly by the Claude Code runtime. What should the developer implement?
- Use a post-tool hook to record outcomes
- Test hook input and exit handling
- Keep hook scripts under change control
- Fail closed when a required safety hook fails
Explanation
Correct Answer
B:Hook contracts specify the event, supplied data, and supported control response; integration tests verify expected enforcement; this behavior directly supports the need to ensure a policy block is interpreted correctly by the Claude Code runtime. A script printing a warning does not necessarily block the tool.
Incorrect Answers
A:This approach serves the separate need to capture which files an approved agent operation actually modified because post-execution events are useful for audit, metrics, and follow-up checks once the tool has run; it does not establish the outcome to ensure a policy block is interpreted correctly by the Claude Code runtime, and they cannot serve as the only prevention mechanism for an irreversible action.
C:For the alternate objective to prevent an unnoticed local edit from disabling a team's protection of secret files, versioning and review make guardrail behavior inspectable and reproducible; the feature here must ensure a policy block is interpreted correctly by the Claude Code runtime, and unreviewed workstation-local scripts can drift across developers.
D:A fail-closed gate treats missing approval as denial for sensitive operations; this applies to the different goal to prevent a tool action from proceeding when its authorization service is unreachable, not the requirement to ensure a policy block is interpreted correctly by the Claude Code runtime; fail-open behavior favors availability at the expense of an important security boundary.
Question 75
An engineer extending a production debugging workflow has been asked to determine why a valid service identity is denied a protected capability. Which choice addresses the request most precisely?
- Investigate a 400 invalid_request error
- Validate data before blaming model quality
- Investigate a 403 authorization failure
- Handle a 429 rate limit error
Explanation
Correct Answer
C:Authorization failures concern permission or policy after identity is established; this behavior directly supports the need to determine why a valid service identity is denied a protected capability. Rotating an otherwise valid key will not necessarily grant missing access.
Incorrect Answers
A:Validation failures point to malformed inputs or unsupported parameters that must be corrected; the supported outcome is to diagnose a request that fails consistently before generating any model output, whereas this scenario requires the ability to determine why a valid service identity is denied a protected capability; backoff does not fix a structurally invalid payload.
B:This approach serves the separate need to diagnose a generated summary that contradicts the latest database balance because source payloads, retrieval filters, and timestamps must be checked before attributing a wrong answer to inference; it does not establish the outcome to determine why a valid service identity is denied a protected capability, and prompt changes cannot repair a stale upstream ledger response.
D:For the alternate objective to prevent a bursty job service from exceeding allowed request throughput, a bounded retry policy and workload shaping address quota or rate throttling; the feature here must determine why a valid service identity is denied a protected capability, and repeating the request immediately can intensify the limit.
Question 76
Two features of a customer platform system are under review. One feature must produce a user-facing answer after a successful inventory lookup; the other must process thousands of nightly documents without interactive deadlines. Which plan matches the respective needs?
- First component: Reconcile batch responses by custom_id. Second component: Submit a Message Batch for offline jobs.
- First component: Send tool_result content back to Claude. Second component: Process tool_use responses in the application.
- First component: Process tool_use responses in the application. Second component: Reconcile batch responses by custom_id.
- First component: Send tool_result content back to Claude. Second component: Submit a Message Batch for offline jobs.
Explanation
Correct Answer
D:The first proposal fits its assigned role because a result block tied to the tool call returns the external observation to the model for synthesis, while the second proposal fits its assigned role because batches suit asynchronous throughput-oriented workloads, with results retrieved after processing; their respective acceptance criteria are to produce a user-facing answer after a successful inventory lookup and to process thousands of nightly documents without interactive deadlines.
Incorrect Answers
A:The first proposal serves the different purpose to match returned batch results to the right source documents despite varying completion order instead of the assigned first role because a caller-assigned identifier links each result to the original request independently of result ordering, while the second proposal fits its assigned role because batches suit asynchronous throughput-oriented workloads, with results retrieved after processing; their respective acceptance criteria are to produce a user-facing answer after a successful inventory lookup and to process thousands of nightly documents without interactive deadlines.
B:The first proposal fits its assigned role because a result block tied to the tool call returns the external observation to the model for synthesis, while the second proposal serves the different purpose to actually retrieve a customer's order status requested by Claude instead of the assigned second role because client tools execute in application code; the model emits structured requests but does not invoke external functions itself; their respective acceptance criteria are to produce a user-facing answer after a successful inventory lookup and to process thousands of nightly documents without interactive deadlines.
C:The first proposal serves the different purpose to actually retrieve a customer's order status requested by Claude instead of the assigned first role because client tools execute in application code; the model emits structured requests but does not invoke external functions itself, while the second proposal serves the different purpose to match returned batch results to the right source documents despite varying completion order instead of the assigned second role because a caller-assigned identifier links each result to the original request independently of result ordering; their respective acceptance criteria are to produce a user-facing answer after a successful inventory lookup and to process thousands of nightly documents without interactive deadlines.
Question 77
A release plan for an agent platform platform has two acceptance checks: component one must coordinate multiple specialists working on interdependent evidence without losing changes; component two must answer questions from changing internal documentation with traceable source excerpts. Which proposal passes both?
- First component: Adopt a blackboard of shared task state. Second component: Summarize long-running state with checkpoints.
- First component: Summarize long-running state with checkpoints. Second component: Use a graph with conditional edges.
- First component: Use a graph with conditional edges. Second component: Use retrieval before generation.
- First component: Adopt a blackboard of shared task state. Second component: Use retrieval before generation.
Explanation
Correct Answer
D:The first proposal fits its assigned role because a controlled shared workspace holds versioned observations and ownership metadata for collaborators, while the second proposal fits its assigned role because retrieval grounds a model answer in current authoritative text rather than hoping parametric knowledge is current; their respective acceptance criteria are to coordinate multiple specialists working on interdependent evidence without losing changes and to answer questions from changing internal documentation with traceable source excerpts.
Incorrect Answers
A:The first proposal fits its assigned role because a controlled shared workspace holds versioned observations and ownership metadata for collaborators, while the second proposal serves the different purpose to continue a support investigation after many rounds of log inspection instead of the assigned second role because periodic compaction preserves decisions, open questions, and evidence references within the context budget; their respective acceptance criteria are to coordinate multiple specialists working on interdependent evidence without losing changes and to answer questions from changing internal documentation with traceable source excerpts.
B:The first proposal serves the different purpose to continue a support investigation after many rounds of log inspection instead of the assigned first role because periodic compaction preserves decisions, open questions, and evidence references within the context budget, while the second proposal serves the different purpose to branch between remediation and escalation based on the result of a security scan instead of the assigned second role because conditional graph transitions encode known branches while allowing model reasoning within selected nodes; their respective acceptance criteria are to coordinate multiple specialists working on interdependent evidence without losing changes and to answer questions from changing internal documentation with traceable source excerpts.
C:The first proposal serves the different purpose to branch between remediation and escalation based on the result of a security scan instead of the assigned first role because conditional graph transitions encode known branches while allowing model reasoning within selected nodes, while the second proposal fits its assigned role because retrieval grounds a model answer in current authoritative text rather than hoping parametric knowledge is current; their respective acceptance criteria are to coordinate multiple specialists working on interdependent evidence without losing changes and to answer questions from changing internal documentation with traceable source excerpts.
Question 78
A conversation design delivery milestone depends on two separate capabilities. One module must obtain named fields in a consistent response layout for subsequent processing; a second module must solve a complex migration comparison that requires several distinct checks. Which implementation option is appropriate?
- First component: Specify output format in the prompt. Second component: Iterate prompts against an evaluation set.
- First component: Specify output format in the prompt. Second component: Use a stepwise decomposition.
- First component: Iterate prompts against an evaluation set. Second component: Specify the handling of missing information.
- First component: Specify the handling of missing information. Second component: Use a stepwise decomposition.
Explanation
Correct Answer
B:The first proposal fits its assigned role because explicit field descriptions guide generation toward the consumer's expected structure, while the second proposal fits its assigned role because breaking the work into verifiable stages lets the application check intermediate artifacts; their respective acceptance criteria are to obtain named fields in a consistent response layout for subsequent processing and to solve a complex migration comparison that requires several distinct checks.
Incorrect Answers
A:The first proposal fits its assigned role because explicit field descriptions guide generation toward the consumer's expected structure, while the second proposal serves the different purpose to improve extraction quality after discovering consistent edge-case failures instead of the assigned second role because controlled revisions measured on stable examples show whether a change actually helps; their respective acceptance criteria are to obtain named fields in a consistent response layout for subsequent processing and to solve a complex migration comparison that requires several distinct checks.
C:The first proposal serves the different purpose to improve extraction quality after discovering consistent edge-case failures instead of the assigned first role because controlled revisions measured on stable examples show whether a change actually helps, while the second proposal serves the different purpose to avoid invented contract dates when the uploaded document lacks them instead of the assigned second role because an explicit unknown or null rule makes uncertainty visible in the requested output; their respective acceptance criteria are to obtain named fields in a consistent response layout for subsequent processing and to solve a complex migration comparison that requires several distinct checks.
D:The first proposal serves the different purpose to avoid invented contract dates when the uploaded document lacks them instead of the assigned first role because an explicit unknown or null rule makes uncertainty visible in the requested output, while the second proposal fits its assigned role because breaking the work into verifiable stages lets the application check intermediate artifacts; their respective acceptance criteria are to obtain named fields in a consistent response layout for subsequent processing and to solve a complex migration comparison that requires several distinct checks.
Question 79
Before a Claude application release, three separately owned tasks must be covered: one service must debug where a request stalls between the web UI, queue, and Claude gateway; another must reject forged callbacks from an external asynchronous job service; and a third must verify that a real staging tool and its API contract work together. Which THREE choices directly address the stated tasks? Choose THREE.
- Validate JSON before deserialization
- Separate integration tests from unit tests
- Refactor duplicated request orchestration
- Correlate distributed traces across services
- Pin dependencies with lockfiles
- Validate webhook signatures
Explanation
Correct Answers
B:Integration tests exercise boundaries and serialization that isolated mocks may miss; this directly supports task 3 to verify that a real staging tool and its API contract work together, not the separate task or tasks to debug where a request stalls between the web UI, queue, and Claude gateway and to reject forged callbacks from an external asynchronous job service; unit tests can pass while authentication or payload mapping is broken.
D:A trace identifier links timing and errors through multiple process boundaries; this directly supports task 1 to debug where a request stalls between the web UI, queue, and Claude gateway, not the separate task or tasks to reject forged callbacks from an external asynchronous job service and to verify that a real staging tool and its API contract work together; local logs with unrelated IDs cannot reconstruct the critical path reliably.
F:Cryptographic verification checks the sender and payload integrity before processing the event; this directly supports task 2 to reject forged callbacks from an external asynchronous job service, not the separate task or tasks to debug where a request stalls between the web UI, queue, and Claude gateway and to verify that a real staging tool and its API contract work together; accepting any POST with the expected JSON fields permits spoofing.
Incorrect Answers
A:A parser and schema validator catch missing types and invalid syntax before side effects; this implements the different outcome to reject a malformed tool argument payload before calling a payroll service rather than the required tasks to verify that a real staging tool and its API contract work together and to reject forged callbacks from an external asynchronous job service and to debug where a request stalls between the web UI, queue, and Claude gateway, and model-generated text is not trustworthy merely because it looks structured, so it is not a direct solution to any of the assigned workstreams.
C:A shared, tested client layer centralizes cross-cutting concerns and reduces inconsistent fixes; this implements the different outcome to maintain multiple Claude entry points without drifting retry and logging behavior rather than the required tasks to verify that a real staging tool and its API contract work together and to reject forged callbacks from an external asynchronous job service and to debug where a request stalls between the web UI, queue, and Claude gateway, and copying the same networking code into every route magnifies defects, so it is not a direct solution to any of the assigned workstreams.
E:Locked versions and hashes constrain drift between build environments; this implements the different outcome to reproduce a working SDK installation across staging and production rather than the required tasks to verify that a real staging tool and its API contract work together and to reject forged callbacks from an external asynchronous job service and to debug where a request stalls between the web UI, queue, and Claude gateway, and unbounded package updates can introduce new behavior during a routine rebuild, so it is not a direct solution to any of the assigned workstreams.
Question 80
A developer is reviewing alternatives for an inference routing feature that must stop a single customer from consuming the entire API allowance during a runaway workflow. Which option is a defensible fit?
- Use batch pricing where deadline permits
- Keep volatile fields after reusable context
- Set budget alerts and tenant caps
- Cache a stable shared prefix
Explanation
Correct Answer
C:Accounting plus enforced limits bounds exposure to unexpected traffic and agent loops; this behavior directly supports the need to stop a single customer from consuming the entire API allowance during a runaway workflow. Usage dashboards alone only report overspend after the fact.
Incorrect Answers
A:Asynchronous batch processing can offer a cost-effective throughput path under supported terms; this applies to the different goal to lower spend for large scheduled classification jobs that do not need immediate responses, not the requirement to stop a single customer from consuming the entire API allowance during a runaway workflow; a night job should not be optimized with live streaming unless its users need it.
B:When an application needs to prevent request timestamps from invalidating a costly static prefix, stable instruction and reference sections should precede changing per-request details for effective prefix reuse; the current capability must instead stop a single customer from consuming the entire API allowance during a runaway workflow, while placing a changing value before the cache boundary undermines matching.
D:This approach serves the separate need to reduce repeated processing of identical tool definitions and system instructions because prefix caching can lower repeated-input cost when the eligible prefix and cache conditions match; it does not establish the outcome to stop a single customer from consuming the entire API allowance during a runaway workflow, and caching does not turn every changed prompt into an automatic hit.