Question 410
Production prompts must be traceable and reproducible across agents and releases. The team is required to record the precise Git revision used for a released agent rather than relying on a moving branch name. Which option is the best fit?
- Prompt rollback commit
- Prompt pull request
- Prompt artifact promotion
- Prompt commit pinning
- Versioned prompt file
- Prompt secret scanning
Explanation
Correct Answer
D:A commit hash fixes the prompt's source state for audit. Unlike a branch tip, it does not move when later commits are merged into the release branch.
Incorrect Answers
A:Reverting the prompt source gives a known prior instruction set; the release process must still redeploy it for live calls to change; the outstanding need is to record the precise Git revision used for a released agent rather than relying on a moving branch name.
B:A pull request provides an approval point for behavioral changes; it does not prove a prompt is safe without relevant tests; this approach leaves the team unable to record the precise Git revision used for a released agent rather than relying on a moving branch name.
C:Artifact promotion avoids a mismatch between evaluated and served instructions; testing one prompt while deploying another invalidates the evaluation evidence; the needed capability is to record the precise Git revision used for a released agent rather than relying on a moving branch name.
E:Source control records exact prompt revisions and supports rollback; it does not automatically update a running agent without deployment automation; the team would still need to record the precise Git revision used for a released agent rather than relying on a moving branch name.
F:Repository scanning reduces accidental disclosure through source history; any already leaked secret still requires rotation and incident review; the specified outcome instead requires the team to record the precise Git revision used for a released agent rather than relying on a moving branch name.
Question 411
Production prompts must be traceable and reproducible across agents and releases. An operator is tasked to require peer review and evaluation results before accepting changes to trusted system instructions. Which mechanism provides that capability?
- Prompt pull request
- Prompt rollback commit
- Prompt commit pinning
- Prompt artifact promotion
Explanation
Correct Answer
A:A pull request provides an approval point for behavioral changes. It does not prove a prompt is safe without relevant tests.
Incorrect Answers
B:Reverting the prompt source gives a known prior instruction set; the release process must still redeploy it for live calls to change; the team would still need to require peer review and evaluation results before accepting changes to trusted system instructions.
C:A commit hash fixes the prompt's source state for audit; unlike a branch tip, it does not move when later commits are merged into the release branch; the specified outcome instead requires the team to require peer review and evaluation results before accepting changes to trusted system instructions.
D:Artifact promotion avoids a mismatch between evaluated and served instructions; testing one prompt while deploying another invalidates the evaluation evidence; the implementation under review needs to require peer review and evaluation results before accepting changes to trusted system instructions.
Question 412
Production prompts must be traceable and reproducible across agents and releases. The immediate design goal is to restore an earlier evaluated prompt when a revision increases unsupported claims in production. Which approach addresses that goal?
- Versioned prompt file
- Prompt rollback commit
- Prompt pull request
- Prompt commit pinning
Explanation
Correct Answer
B:Reverting the prompt source gives a known prior instruction set. The release process must still redeploy it for live calls to change.
Incorrect Answers
A:Source control records exact prompt revisions and supports rollback; it does not automatically update a running agent without deployment automation; the implementation under review needs to restore an earlier evaluated prompt when a revision increases unsupported claims in production.
C:A pull request provides an approval point for behavioral changes; it does not prove a prompt is safe without relevant tests; this alternative does not achieve the requirement to restore an earlier evaluated prompt when a revision increases unsupported claims in production.
D:A commit hash fixes the prompt's source state for audit; unlike a branch tip, it does not move when later commits are merged into the release branch; the actual design criterion is to restore an earlier evaluated prompt when a revision increases unsupported claims in production.
Question 413
Production prompts must be traceable and reproducible across agents and releases. The team is required to publish the tested prompt revision alongside model and retrieval configuration in the deployment manifest. Which option is the best fit?
- Prompt pull request
- Prompt secret scanning
- Prompt commit pinning
- Prompt artifact promotion
- Versioned prompt file
Explanation
Correct Answer
D:Artifact promotion avoids a mismatch between evaluated and served instructions. Testing one prompt while deploying another invalidates the evaluation evidence.
Incorrect Answers
A:A pull request provides an approval point for behavioral changes; it does not prove a prompt is safe without relevant tests; the actual design criterion is to publish the tested prompt revision alongside model and retrieval configuration in the deployment manifest.
B:Repository scanning reduces accidental disclosure through source history; any already leaked secret still requires rotation and incident review; the outstanding need is to publish the tested prompt revision alongside model and retrieval configuration in the deployment manifest.
C:A commit hash fixes the prompt's source state for audit; unlike a branch tip, it does not move when later commits are merged into the release branch; this approach leaves the team unable to publish the tested prompt revision alongside model and retrieval configuration in the deployment manifest.
E:Source control records exact prompt revisions and supports rollback; it does not automatically update a running agent without deployment automation; the requirement, by contrast, is to publish the tested prompt revision alongside model and retrieval configuration in the deployment manifest.
Question 414
Production prompts must be traceable and reproducible across agents and releases. An operator is tasked to detect accidentally embedded credentials or private customer data before a prompt file is merged. Which mechanism provides that capability?
- Prompt secret scanning
- Versioned prompt file
- Prompt commit pinning
- Prompt rollback commit
Explanation
Correct Answer
A:Repository scanning reduces accidental disclosure through source history. Any already leaked secret still requires rotation and incident review.
Incorrect Answers
B:Source control records exact prompt revisions and supports rollback; it does not automatically update a running agent without deployment automation; the requirement, by contrast, is to detect accidentally embedded credentials or private customer data before a prompt file is merged.
C:A commit hash fixes the prompt's source state for audit; unlike a branch tip, it does not move when later commits are merged into the release branch; the team would still need to detect accidentally embedded credentials or private customer data before a prompt file is merged.
D:Reverting the prompt source gives a known prior instruction set; the release process must still redeploy it for live calls to change; the specified outcome instead requires the team to detect accidentally embedded credentials or private customer data before a prompt file is merged.
Question 415
A DevOps team integrates prompt and application changes into controlled CI/CD. The immediate design goal is to reject a template with missing variables or invalid structured-output configuration before integration tests run. Which approach addresses that goal?
- Threshold-based pipeline failure
- Model-prompt compatibility check
- Automated evaluation job
- Prompt syntax validation
Explanation
Correct Answer
D:Static validation catches deterministic template errors early. It cannot assess whether a well-formed prompt produces correct answers.
Incorrect Answers
A:An enforced threshold prevents accidental promotion of known regressions; a dashboard warning without pipeline enforcement is weaker release protection; the specified outcome instead requires the team to reject a template with missing variables or invalid structured-output configuration before integration tests run.
B:Compatibility tests catch changes in model interfaces or output shape; a prompt that works against one provider endpoint may fail against another; the implementation under review needs to reject a template with missing variables or invalid structured-output configuration before integration tests run.
C:Evaluation jobs generate comparable quality results for a proposed revision; they should use pinned model and dataset versions to limit confounding changes; the requested task remains to reject a template with missing variables or invalid structured-output configuration before integration tests run.
Question 416
A DevOps team integrates prompt and application changes into controlled CI/CD. The team is required to run a stable agent or prompt evaluation dataset as part of each candidate release pipeline. Which option is the best fit?
- Automated evaluation job
- Post-release smoke test
- Threshold-based pipeline failure
- Prompt syntax validation
Explanation
Correct Answer
A:Evaluation jobs generate comparable quality results for a proposed revision. They should use pinned model and dataset versions to limit confounding changes.
Incorrect Answers
B:Smoke tests catch wiring and credential defects after promotion; they complement deeper offline evaluations but cannot establish comprehensive quality; the actual design criterion is to run a stable agent or prompt evaluation dataset as part of each candidate release pipeline.
C:An enforced threshold prevents accidental promotion of known regressions; a dashboard warning without pipeline enforcement is weaker release protection; the outstanding need is to run a stable agent or prompt evaluation dataset as part of each candidate release pipeline.
D:Static validation catches deterministic template errors early; it cannot assess whether a well-formed prompt produces correct answers; this approach leaves the team unable to run a stable agent or prompt evaluation dataset as part of each candidate release pipeline.
Question 417
A DevOps team integrates prompt and application changes into controlled CI/CD. An operator is tasked to fail CI when safety or task-quality scores drop below explicitly approved acceptance values. Which mechanism provides that capability?
- Threshold-based pipeline failure
- Post-release smoke test
- Model-prompt compatibility check
- Prompt syntax validation
Explanation
Correct Answer
A:An enforced threshold prevents accidental promotion of known regressions. A dashboard warning without pipeline enforcement is weaker release protection.
Incorrect Answers
B:Smoke tests catch wiring and credential defects after promotion; they complement deeper offline evaluations but cannot establish comprehensive quality; the needed capability is to fail CI when safety or task-quality scores drop below explicitly approved acceptance values.
C:Compatibility tests catch changes in model interfaces or output shape; a prompt that works against one provider endpoint may fail against another; the requirement, by contrast, is to fail CI when safety or task-quality scores drop below explicitly approved acceptance values.
D:Static validation catches deterministic template errors early; it cannot assess whether a well-formed prompt produces correct answers; the team would still need to fail CI when safety or task-quality scores drop below explicitly approved acceptance values.
Question 418
A DevOps team integrates prompt and application changes into controlled CI/CD. The immediate design goal is to require production-owner authorization after CI checks pass for a behavior-changing agent update. Which approach addresses that goal?
- Deployment approval
- Model-prompt compatibility check
- Automated evaluation job
- Threshold-based pipeline failure
Explanation
Correct Answer
A:Approvals create accountability for accepted residual risk. A passing automated test does not eliminate the need for domain-specific oversight.
Incorrect Answers
B:Compatibility tests catch changes in model interfaces or output shape; a prompt that works against one provider endpoint may fail against another; the specified outcome instead requires the team to require production-owner authorization after CI checks pass for a behavior-changing agent update.
C:Evaluation jobs generate comparable quality results for a proposed revision; they should use pinned model and dataset versions to limit confounding changes; the implementation under review needs to require production-owner authorization after CI checks pass for a behavior-changing agent update.
D:An enforced threshold prevents accidental promotion of known regressions; a dashboard warning without pipeline enforcement is weaker release protection; the requested task remains to require production-owner authorization after CI checks pass for a behavior-changing agent update.
Question 419
A DevOps team integrates prompt and application changes into controlled CI/CD. The team is required to validate a prompt's tool syntax and response parser against the exact target deployment version. Which option is the best fit?
- Model-prompt compatibility check
- Deployment approval
- Automated evaluation job
- Post-release smoke test
Explanation
Correct Answer
A:Compatibility tests catch changes in model interfaces or output shape. A prompt that works against one provider endpoint may fail against another.
Incorrect Answers
B:Approvals create accountability for accepted residual risk; a passing automated test does not eliminate the need for domain-specific oversight; this alternative does not achieve the requirement to validate a prompt's tool syntax and response parser against the exact target deployment version.
C:Evaluation jobs generate comparable quality results for a proposed revision; they should use pinned model and dataset versions to limit confounding changes; the actual design criterion is to validate a prompt's tool syntax and response parser against the exact target deployment version.
D:Smoke tests catch wiring and credential defects after promotion; they complement deeper offline evaluations but cannot establish comprehensive quality; the outstanding need is to validate a prompt's tool syntax and response parser against the exact target deployment version.
Question 420
A prompt-update pipeline already validates syntax, evaluates thresholds, and requires release-owner approval. An incident shows that a change passed those gates but was never exercised against realistic test queries during CI or checked after deployment. Which TWO automated activities add those missing pre-release and post-release checks? Choose TWO.
- Prompt syntax validation
- Threshold-based pipeline failure
- Post-release smoke test
- Automated evaluation job
- Deployment approval
Explanation
Correct Answers
C:Smoke tests catch wiring and credential defects after promotion; they complement deeper offline evaluations but cannot establish comprehensive quality. Here, the new endpoint is checked after promotion rather than relying solely on pipeline success.
D:Evaluation jobs generate comparable quality results for a proposed revision; they should use pinned model and dataset versions to limit confounding changes. Here, a stable representative test set exercises the changed behavior before release.
Incorrect Answers
A:Static validation catches deterministic template errors early; it cannot assess whether a well-formed prompt produces correct answers; it does not by itself address representative CI behavioral evaluation and a postdeployment functional check.
B:An enforced threshold prevents accidental promotion of known regressions; a dashboard warning without pipeline enforcement is weaker release protection; the observed fault still involves representative CI behavioral evaluation and a postdeployment functional check.
E:Approvals create accountability for accepted residual risk; a passing automated test does not eliminate the need for domain-specific oversight; this mechanism leaves unresolved representative CI behavioral evaluation and a postdeployment functional check.