Anthropic CCDV-F (Claude Certified Developer - Foundations) Practice Test 55

Question 540

A backend integration review requires two independent, explicitly defined outcomes: the first workstream must trace a quality regression to the exact instruction revision used; a second must keep external tool connectors compatible after client upgrades. Which TWO actions directly provide these capabilities? Choose TWO.

  1. Track MCP server versions and contracts
  2. Put repository guidance in CLAUDE.md
  3. Use settings.json for Claude Code permissions
  4. Keep environment-specific settings external
  5. Version prompt templates alongside code

Explanation

Correct Answers

A:

Declared versions and integration tests expose resource or schema changes before deployment; this directly supports task 2 to keep external tool connectors compatible after client upgrades, not the separate task or tasks to trace a quality regression to the exact instruction revision used; automatically using the newest connector can alter behavior unexpectedly.

E:

Source-controlled prompt assets make rollbacks, diffs, and evaluations possible; this directly supports task 1 to trace a quality regression to the exact instruction revision used, not the separate task or tasks to keep external tool connectors compatible after client upgrades; editing prompts in an unversioned dashboard hides causal history.

Incorrect Answers

B:

Repository memory files provide persistent instructional context to coding sessions; this implements the different outcome to give Claude Code the project's build, test, and architecture conventions rather than the required tasks to trace a quality regression to the exact instruction revision used and to keep external tool connectors compatible after client upgrades, and they are not an application secrets manager or the API's global system prompt, so it is not a direct solution to any of the assigned workstreams.

C:

Permission settings define configurable controls for Claude Code rather than merely describing preferences; this implements the different outcome to constrain a repository coding assistant's allowed tools and execution behavior rather than the required tasks to trace a quality regression to the exact instruction revision used and to keep external tool connectors compatible after client upgrades, and a CLAUDE.md comment cannot be relied on as a hard execution gate, so it is not a direct solution to any of the assigned workstreams.

D:

Configuration injection separates environment differences from immutable application artifacts; this implements the different outcome to deploy the same application build to staging and production with distinct endpoints rather than the required tasks to trace a quality regression to the exact instruction revision used and to keep external tool connectors compatible after client upgrades, and hard-coded URLs require risky code changes for promotion, so it is not a direct solution to any of the assigned workstreams.

Question 541

Before an external system access release, three separately owned tasks must be covered: one service must require stricter approval for actions that change customer data; another must stop a directory lookup function from also gaining unrestricted database write access; and a third must prevent a helpful assistant from issuing refunds automatically. Which THREE choices directly address the stated tasks? Choose THREE.

  1. Validate generated arguments in application code
  2. Separate read tools from write tools
  3. Write task-specific tool descriptions
  4. Add an approval boundary before sensitive writes
  5. Restrict tool network credentials
  6. Return structured tool errors

Explanation

Correct Answers

B:

Distinct interfaces simplify least-privilege grants and side-effect controls; this directly supports task 1 to require stricter approval for actions that change customer data, not the separate task or tasks to stop a directory lookup function from also gaining unrestricted database write access and to prevent a helpful assistant from issuing refunds automatically; a generic execute-anything function makes authorization difficult to enforce.

D:

Authorization by a user or operator can be enforced before the executor commits changes; this directly supports task 3 to prevent a helpful assistant from issuing refunds automatically, not the separate task or tasks to require stricter approval for actions that change customer data and to stop a directory lookup function from also gaining unrestricted database write access; a note in the model's prompt is not a transaction approval.

E:

Minimal capabilities reduce risk when generated arguments or retrieved text are hostile; this directly supports task 2 to stop a directory lookup function from also gaining unrestricted database write access, not the separate task or tasks to require stricter approval for actions that change customer data and to prevent a helpful assistant from issuing refunds automatically; better prompt wording does not remove excess infrastructure permissions.

Incorrect Answers

A:

The executor must check types, ranges, identity, and scope independently of the model; this implements the different outcome to prevent malformed or unauthorized input from executing a customer data query rather than the required tasks to require stricter approval for actions that change customer data and to prevent a helpful assistant from issuing refunds automatically and to stop a directory lookup function from also gaining unrestricted database write access, and a valid tool schema by itself is not a complete authorization system, so it is not a direct solution to any of the assigned workstreams.

C:

Clear capabilities and side effects guide tool selection without granting new authority; this implements the different outcome to let Claude distinguish a read-only invoice lookup from an invoice update rather than the required tasks to require stricter approval for actions that change customer data and to prevent a helpful assistant from issuing refunds automatically and to stop a directory lookup function from also gaining unrestricted database write access, and vague overlapping descriptions increase erroneous calls, so it is not a direct solution to any of the assigned workstreams.

F:

Machine-readable failures distinguish not-found, permission denied, and transient errors for appropriate follow-up; this implements the different outcome to let an assistant recover from a missing order without inventing its details rather than the required tasks to require stricter approval for actions that change customer data and to prevent a helpful assistant from issuing refunds automatically and to stop a directory lookup function from also gaining unrestricted database write access, and pretending failure is an empty successful record hides operational meaning, so it is not a direct solution to any of the assigned workstreams.

Question 542

A source-control workflow build contains an upstream function and a separate downstream function. The upstream work must deny destructive commands regardless of what a generated plan proposes; the downstream work must continue an interrupted refactor while retaining relevant prior decisions. Which configuration pair covers both?

  1. First component: Run repository initialization to create project guidance. Second component: Create a reusable Skill.
  2. First component: Connect an MCP server to Claude Code. Second component: Run repository initialization to create project guidance.
  3. First component: Create a reusable Skill. Second component: Review and resume a saved session.
  4. First component: Configure permissions in Claude Code settings. Second component: Review and resume a saved session.
  5. First component: Configure permissions in Claude Code settings. Second component: Run repository initialization to create project guidance.

Explanation

Correct Answer

D:

The first proposal fits its assigned role because explicit allow and deny rules establish executable tool policy rather than advisory prose, while the second proposal fits its assigned role because session continuation can reestablish conversational state without restarting the task from scratch; their respective acceptance criteria are to deny destructive commands regardless of what a generated plan proposes and to continue an interrupted refactor while retaining relevant prior decisions.

Incorrect Answers

A:

The first proposal serves the different purpose to establish a discoverable set of build and test conventions for a new Claude Code workspace instead of the assigned first role because repository initialization can generate or prepare project instruction material that the team subsequently reviews, while the second proposal serves the different purpose to make a specialised review procedure available as task-specific instructions and assets instead of the assigned second role because a Skill packages a repeatable capability that Claude can load when relevant to the work; their respective acceptance criteria are to deny destructive commands regardless of what a generated plan proposes and to continue an interrupted refactor while retaining relevant prior decisions.

B:

The first proposal serves the different purpose to expose approved project-specific actions without copying credentials into prompts instead of the assigned first role because an MCP integration declares external capabilities and their transport while access remains controlled by the server, while the second proposal serves the different purpose to establish a discoverable set of build and test conventions for a new Claude Code workspace instead of the assigned second role because repository initialization can generate or prepare project instruction material that the team subsequently reviews; their respective acceptance criteria are to deny destructive commands regardless of what a generated plan proposes and to continue an interrupted refactor while retaining relevant prior decisions.

C:

The first proposal serves the different purpose to make a specialised review procedure available as task-specific instructions and assets instead of the assigned first role because a Skill packages a repeatable capability that Claude can load when relevant to the work, while the second proposal fits its assigned role because session continuation can reestablish conversational state without restarting the task from scratch; their respective acceptance criteria are to deny destructive commands regardless of what a generated plan proposes and to continue an interrupted refactor while retaining relevant prior decisions.

E:

The first proposal fits its assigned role because explicit allow and deny rules establish executable tool policy rather than advisory prose, while the second proposal serves the different purpose to establish a discoverable set of build and test conventions for a new Claude Code workspace instead of the assigned second role because repository initialization can generate or prepare project instruction material that the team subsequently reviews; their respective acceptance criteria are to deny destructive commands regardless of what a generated plan proposes and to continue an interrupted refactor while retaining relevant prior decisions.

Question 543

An agent platform review requires two independent, explicitly defined outcomes: the first workstream must build a coding agent that needs managed tool execution, sessions, and event handling; a second must recover from a failed search while preserving the agent's chance to choose another path. Which TWO actions directly provide these capabilities? Choose TWO.

  1. Set a tool execution timeout
  2. Execute independent requested tools concurrently
  3. Return a structured tool error to the model
  4. Use the Claude Agent SDK for an integrated tool-driven agent
  5. Choose isolated sandbox execution for untrusted code

Explanation

Correct Answers

C:

Tool-result errors surface the failure without impersonating a successful action; this directly supports task 2 to recover from a failed search while preserving the agent's chance to choose another path, not the separate task or tasks to build a coding agent that needs managed tool execution, sessions, and event handling; silently dropping a failed tool response leaves the model without a valid observation.

D:

The SDK provides agent-loop primitives and integrations that remove much of the custom harness work; this directly supports task 1 to build a coding agent that needs managed tool execution, sessions, and event handling, not the separate task or tasks to recover from a failed search while preserving the agent's chance to choose another path; it still requires explicit permissions, safety policies, and operational boundaries.

Incorrect Answers

A:

Deadlines bound latency and convert stalled calls into explicit recoverable failures; this implements the different outcome to prevent a slow external inventory service from blocking an agent indefinitely rather than the required tasks to build a coding agent that needs managed tool execution, sessions, and event handling and to recover from a failed search while preserving the agent's chance to choose another path, and a generous model output limit cannot constrain external network operations, so it is not a direct solution to any of the assigned workstreams.

B:

Parallel execution is safe for independent operations while every result must still be matched to its call ID; this implements the different outcome to reduce turnaround when an agent asks for unrelated reads in the same response rather than the required tasks to build a coding agent that needs managed tool execution, sessions, and event handling and to recover from a failed search while preserving the agent's chance to choose another path, and mutating tools with data dependencies should not be blindly parallelized, so it is not a direct solution to any of the assigned workstreams.

E:

A restricted execution environment limits filesystem, network, and resource privileges; this implements the different outcome to let an agent run generated scripts without exposing production host files rather than the required tasks to build a coding agent that needs managed tool execution, sessions, and event handling and to recover from a failed search while preserving the agent's chance to choose another path, and running code in the main application process expands the blast radius of a bad tool call, so it is not a direct solution to any of the assigned workstreams.

Question 544

A backend integration team is evaluating two disjoint use cases. For one use case, it must provision an interactive application for peak seasonal traffic; for the other, it must turn a vague request for an AI claims assistant into observable user actions and outputs. Which paired choices are justified?

  1. First component: Set an explicit data residency constraint. Second component: Document functional acceptance criteria.
  2. First component: Calculate expected traffic and concurrency. Second component: Record nonfunctional latency targets.
  3. First component: Record nonfunctional latency targets. Second component: Set an explicit data residency constraint.
  4. First component: Calculate expected traffic and concurrency. Second component: Document functional acceptance criteria.

Explanation

Correct Answer

D:

The first proposal fits its assigned role because workload assumptions drive rate-limit planning, queue capacity, and retry strategy, while the second proposal fits its assigned role because acceptance criteria specify inputs, outputs, and successful behavior that can be tested before release; their respective acceptance criteria are to provision an interactive application for peak seasonal traffic and to turn a vague request for an AI claims assistant into observable user actions and outputs.

Incorrect Answers

A:

The first proposal serves the different purpose to ensure protected records are processed only in approved jurisdictions instead of the assigned first role because documented residency requirements inform cloud placement, logging, storage, and vendor selection, while the second proposal fits its assigned role because acceptance criteria specify inputs, outputs, and successful behavior that can be tested before release; their respective acceptance criteria are to provision an interactive application for peak seasonal traffic and to turn a vague request for an AI claims assistant into observable user actions and outputs.

B:

The first proposal fits its assigned role because workload assumptions drive rate-limit planning, queue capacity, and retry strategy, while the second proposal serves the different purpose to decide between streaming and offline processing for a time-sensitive support assistant instead of the assigned second role because a measurable response-time objective constrains interaction design and infrastructure choices; their respective acceptance criteria are to provision an interactive application for peak seasonal traffic and to turn a vague request for an AI claims assistant into observable user actions and outputs.

C:

The first proposal serves the different purpose to decide between streaming and offline processing for a time-sensitive support assistant instead of the assigned first role because a measurable response-time objective constrains interaction design and infrastructure choices, while the second proposal serves the different purpose to ensure protected records are processed only in approved jurisdictions instead of the assigned second role because documented residency requirements inform cloud placement, logging, storage, and vendor selection; their respective acceptance criteria are to provision an interactive application for peak seasonal traffic and to turn a vague request for an AI claims assistant into observable user actions and outputs.

Question 545

An incident response review requires two independent, explicitly defined outcomes: the first workstream must retain complete control over a small set of proprietary tool calls and their execution order; a second must avoid charging a customer twice after a worker retries a timed-out payment action. Which TWO actions directly provide these capabilities? Choose TWO.

  1. Implement a custom Messages API tool loop
  2. Use the Claude Agent SDK for an integrated tool-driven agent
  3. Keep user-specific state out of a shared agent process
  4. Add idempotency keys to side-effecting operations
  5. Forward tool_result blocks with matching identifiers

Explanation

Correct Answers

A:

The application receives tool requests, validates inputs, executes authorized functions, and sends tool results back; this directly supports task 1 to retain complete control over a small set of proprietary tool calls and their execution order, not the separate task or tasks to avoid charging a customer twice after a worker retries a timed-out payment action; the Messages API does not automatically execute arbitrary client tools.

D:

Application-generated operation identifiers allow the downstream service to recognize a repeated request; this directly supports task 2 to avoid charging a customer twice after a worker retries a timed-out payment action, not the separate task or tasks to retain complete control over a small set of proprietary tool calls and their execution order; model-level instructions cannot guarantee exactly-once execution after a partial failure.

Incorrect Answers

B:

The SDK provides agent-loop primitives and integrations that remove much of the custom harness work; this implements the different outcome to build a coding agent that needs managed tool execution, sessions, and event handling rather than the required tasks to retain complete control over a small set of proprietary tool calls and their execution order and to avoid charging a customer twice after a worker retries a timed-out payment action, and it still requires explicit permissions, safety policies, and operational boundaries, so it is not a direct solution to any of the assigned workstreams.

C:

Per-request identity and session boundaries prevent cross-tenant context leakage; this implements the different outcome to serve many tenants without mixing their tool outputs or prior messages rather than the required tasks to retain complete control over a small set of proprietary tool calls and their execution order and to avoid charging a customer twice after a worker retries a timed-out payment action, and shared conversation histories can inadvertently expose unrelated customers' data, so it is not a direct solution to any of the assigned workstreams.

E:

Results must reference the corresponding tool_use IDs and be included in the next conversation turn; this implements the different outcome to let Claude continue after a client tool returns a transaction status rather than the required tasks to retain complete control over a small set of proprietary tool calls and their execution order and to avoid charging a customer twice after a worker retries a timed-out payment action, and sending a human-readable summary without the tool correlation can break the protocol, so it is not a direct solution to any of the assigned workstreams.

Question 546

After pilot feedback, an agent governance group needs two independent improvements. The first is to display generated customer notes safely in a browser; the second is to block a generated database query that targets another tenant's records. Which engineering proposal handles both?

  1. First component: Apply retention limits to diagnostic logs. Second component: Scope database credentials by tenant and operation.
  2. First component: Use output escaping to prevent XSS. Second component: Apply retention limits to diagnostic logs.
  3. First component: Use output escaping to prevent XSS. Second component: Validate tool arguments server-side.
  4. First component: Scope database credentials by tenant and operation. Second component: Validate tool arguments server-side.

Explanation

Correct Answer

C:

The first proposal fits its assigned role because context-aware encoding prevents text from being interpreted as active HTML or JavaScript, while the second proposal fits its assigned role because backend authorization checks must bind the requested operation to the caller's entitlements; their respective acceptance criteria are to display generated customer notes safely in a browser and to block a generated database query that targets another tenant's records.

Incorrect Answers

A:

The first proposal serves the different purpose to reduce exposure of customer documents copied into application telemetry instead of the assigned first role because data minimization and controlled retention reduce the duration and audience of sensitive log content, while the second proposal serves the different purpose to prevent a read-only support tool from modifying billing tables instead of the assigned second role because least-privilege identity limits the damage from a manipulated tool call; their respective acceptance criteria are to display generated customer notes safely in a browser and to block a generated database query that targets another tenant's records.

B:

The first proposal fits its assigned role because context-aware encoding prevents text from being interpreted as active HTML or JavaScript, while the second proposal serves the different purpose to reduce exposure of customer documents copied into application telemetry instead of the assigned second role because data minimization and controlled retention reduce the duration and audience of sensitive log content; their respective acceptance criteria are to display generated customer notes safely in a browser and to block a generated database query that targets another tenant's records.

D:

The first proposal serves the different purpose to prevent a read-only support tool from modifying billing tables instead of the assigned first role because least-privilege identity limits the damage from a manipulated tool call, while the second proposal fits its assigned role because backend authorization checks must bind the requested operation to the caller's entitlements; their respective acceptance criteria are to display generated customer notes safely in a browser and to block a generated database query that targets another tenant's records.

Question 547

During a latency optimization rollout, two different failure paths need remediation. The first component must implement an interactive collaboration channel requiring messages in both directions; separately, the second must deliver tokens to a browser without maintaining a bidirectional command channel. Which changes should be adopted?

  1. First component: Use the Messages count_tokens capability. Second component: Choose SSE for one-way streamed responses.
  2. First component: Use direct HTTP REST integration for unsupported runtimes. Second component: Use the Messages count_tokens capability.
  3. First component: Choose WebSockets for bidirectional realtime exchange. Second component: Choose SSE for one-way streamed responses.
  4. First component: Choose WebSockets for bidirectional realtime exchange. Second component: Use direct HTTP REST integration for unsupported runtimes.

Explanation

Correct Answer

C:

The first proposal fits its assigned role because persistent two-way channels fit interactive event exchange where the client also sends messages during a session, while the second proposal fits its assigned role because server-sent events provide ordered server-to-client updates over an HTTP stream; their respective acceptance criteria are to implement an interactive collaboration channel requiring messages in both directions and to deliver tokens to a browser without maintaining a bidirectional command channel.

Incorrect Answers

A:

The first proposal serves the different purpose to estimate model-token input demand before launching a large processing job instead of the assigned first role because token counting helps check input budgets before generation and cost planning, while the second proposal fits its assigned role because server-sent events provide ordered server-to-client updates over an HTTP stream; their respective acceptance criteria are to implement an interactive collaboration channel requiring messages in both directions and to deliver tokens to a browser without maintaining a bidirectional command channel.

B:

The first proposal serves the different purpose to call the Messages endpoint from a language without a mature provider SDK instead of the assigned first role because a standards-based HTTP client can send JSON requests with the correct headers and decode responses, while the second proposal serves the different purpose to estimate model-token input demand before launching a large processing job instead of the assigned second role because token counting helps check input budgets before generation and cost planning; their respective acceptance criteria are to implement an interactive collaboration channel requiring messages in both directions and to deliver tokens to a browser without maintaining a bidirectional command channel.

D:

The first proposal fits its assigned role because persistent two-way channels fit interactive event exchange where the client also sends messages during a session, while the second proposal serves the different purpose to call the Messages endpoint from a language without a mature provider SDK instead of the assigned second role because a standards-based HTTP client can send JSON requests with the correct headers and decode responses; their respective acceptance criteria are to implement an interactive collaboration channel requiring messages in both directions and to deliver tokens to a browser without maintaining a bidirectional command channel.

Question 548

After pilot feedback, an incident response group needs two independent improvements. The first is to resume a multi-step agentic process after worker or network failure; the second is to investigate an open-ended incident where the next query depends on earlier findings. Which engineering proposal handles both?

  1. First component: Use a central state record for orchestration. Second component: Use an agentic decision loop.
  2. First component: Apply routing before task execution. Second component: Use an agentic decision loop.
  3. First component: Use sequential chaining for dependent subtasks. Second component: Apply routing before task execution.
  4. First component: Use a central state record for orchestration. Second component: Use sequential chaining for dependent subtasks.

Explanation

Correct Answer

A:

The first proposal fits its assigned role because persisted job state tracks completed stages, pending actions, and recoverable checkpoints outside the model context, while the second proposal fits its assigned role because a model-directed loop can choose subsequent actions based on tool observations until it reaches a stopping condition; their respective acceptance criteria are to resume a multi-step agentic process after worker or network failure and to investigate an open-ended incident where the next query depends on earlier findings.

Incorrect Answers

B:

The first proposal serves the different purpose to choose the right specialised handling path from the request category instead of the assigned first role because an inexpensive classification or rule-based router sends each request to the applicable predefined workflow, while the second proposal fits its assigned role because a model-directed loop can choose subsequent actions based on tool observations until it reaches a stopping condition; their respective acceptance criteria are to resume a multi-step agentic process after worker or network failure and to investigate an open-ended incident where the next query depends on earlier findings.

C:

The first proposal serves the different purpose to ensure a generated specification is validated before code generation begins instead of the assigned first role because each stage consumes an approved artifact from the previous stage, preserving dependency order and stage-specific checks, while the second proposal serves the different purpose to choose the right specialised handling path from the request category instead of the assigned second role because an inexpensive classification or rule-based router sends each request to the applicable predefined workflow; their respective acceptance criteria are to resume a multi-step agentic process after worker or network failure and to investigate an open-ended incident where the next query depends on earlier findings.

D:

The first proposal fits its assigned role because persisted job state tracks completed stages, pending actions, and recoverable checkpoints outside the model context, while the second proposal serves the different purpose to ensure a generated specification is validated before code generation begins instead of the assigned second role because each stage consumes an approved artifact from the previous stage, preserving dependency order and stage-specific checks; their respective acceptance criteria are to resume a multi-step agentic process after worker or network failure and to investigate an open-ended incident where the next query depends on earlier findings.

Question 549

A retrieval pipeline application is approaching release, but the team still needs to obtain named fields in a consistent response layout for subsequent processing. What is the most appropriate implementation?

  1. Specify the handling of missing information
  2. State one measurable outcome per instruction
  3. Specify output format in the prompt
  4. Iterate prompts against an evaluation set

Explanation

Correct Answer

C:

Explicit field descriptions guide generation toward the consumer's expected structure; this behavior directly supports the need to obtain named fields in a consistent response layout for subsequent processing. Formatting guidance still requires machine validation at the boundary.

Incorrect Answers

A:

When an application needs to avoid invented contract dates when the uploaded document lacks them, an explicit unknown or null rule makes uncertainty visible in the requested output; the current capability must instead obtain named fields in a consistent response layout for subsequent processing, while a strict field list is not permission to fabricate absent values.

B:

Clear task goals and acceptance criteria reduce competing interpretations; the supported outcome is to stop a summarizer from producing unrelated analysis instead of required action items, whereas this scenario requires the ability to obtain named fields in a consistent response layout for subsequent processing; a vague request for a good summary provides no testable target.

D:

Controlled revisions measured on stable examples show whether a change actually helps; despite being useful when a service must improve extraction quality after discovering consistent edge-case failures, it is not a solution to obtain named fields in a consistent response layout for subsequent processing; optimizing against a single anecdote can degrade other cases.

Question 550

Engineers are splitting a data pipeline service into two independently tested functions. Function A must reject a malformed tool argument payload before calling a payroll service, while function B must activate a new tool execution path for only internal testers. Which plan is sound?

  1. First component: Validate JSON before deserialization. Second component: Use feature flags for risky behavior changes.
  2. First component: Separate integration tests from unit tests. Second component: Use pull requests and focused code review.
  3. First component: Use pull requests and focused code review. Second component: Use feature flags for risky behavior changes.
  4. First component: Validate JSON before deserialization. Second component: Separate integration tests from unit tests.

Explanation

Correct Answer

A:

The first proposal fits its assigned role because a parser and schema validator catch missing types and invalid syntax before side effects, while the second proposal fits its assigned role because runtime controls can segment rollout and permit quick disabling without a code rebuild; their respective acceptance criteria are to reject a malformed tool argument payload before calling a payroll service and to activate a new tool execution path for only internal testers.

Incorrect Answers

B:

The first proposal serves the different purpose to verify that a real staging tool and its API contract work together instead of the assigned first role because integration tests exercise boundaries and serialization that isolated mocks may miss, while the second proposal serves the different purpose to detect unsafe permissions and schema changes before merging an agent tool instead of the assigned second role because peer review creates an explicit checkpoint for correctness, security, and maintainability; their respective acceptance criteria are to reject a malformed tool argument payload before calling a payroll service and to activate a new tool execution path for only internal testers.

C:

The first proposal serves the different purpose to detect unsafe permissions and schema changes before merging an agent tool instead of the assigned first role because peer review creates an explicit checkpoint for correctness, security, and maintainability, while the second proposal fits its assigned role because runtime controls can segment rollout and permit quick disabling without a code rebuild; their respective acceptance criteria are to reject a malformed tool argument payload before calling a payroll service and to activate a new tool execution path for only internal testers.

D:

The first proposal fits its assigned role because a parser and schema validator catch missing types and invalid syntax before side effects, while the second proposal serves the different purpose to verify that a real staging tool and its API contract work together instead of the assigned second role because integration tests exercise boundaries and serialization that isolated mocks may miss; their respective acceptance criteria are to reject a malformed tool argument payload before calling a payroll service and to activate a new tool execution path for only internal testers.

Claude Certified Developer - Foundations (CCDV-F) Practice Tests