Your document extraction tool uses ML models to extract invoice fields (vendor, amount, date).
The models return confidence scores (0.0-1.0) for each extracted field. In production, you observe: (1) the agent proceeds with low-confidence extractions that are incorrect 23% of the time, and (2) the agent requests unnecessary human review for 31% of extractions that were actually correct. How should you restructure the tool's output?

• A. Return fields with their raw confidence scores and add detailed few-shot examples to your system prompt demonstrating how to interpret different confidence ranges and when to request human review.
• B. Return fields organized into verified and needs_verification objects based on confidence thresholds.
• C. Compute an aggregate extraction_quality score across all fields and return it alongside the extracted values. Include a text summary describing the overall extraction reliability.
• D. Return fields with confidence scores, plus a request_review boolean computed using your tested confidence thresholds, along with a review_reasons array explaining which fields triggered review.

Users report that final reports sometimes lack depth on specific subtopics. Investigation shows that the document analysis agent frequently identifies gaps - for instance, noting "the retrieved sources discuss API authentication but lack details on token refresh patterns" - but under the current strict pipeline, this insight isn't actionable since search has already completed. What's the most effective architectural change?

• A. Add a research planning agent before the search phase that decomposes topics into specific sub- questions.
• B. Have the coordinator review analysis output for gap indicators and re-invoke search with gap- informed queries when gaps are detected.
• C. Have the analysis agent report specific gaps to the coordinator, which triggers targeted searches and re-invokes analysis until sufficient.
• D. Have the synthesis agent attach confidence scores to each section and flag areas with insufficient coverage for manual review.

Your automated review generates many findings per pull request, but developer feedback shows that roughly half are dismissed as "not worth addressing." Analysis reveals that dismissed findings are often technically accurate but involve minor style preferences or patterns that are acceptable in your codebase. Before adding infrastructure complexity, what prompt-design change could most effectively reduce dismissals while maintaining the detection of genuine issues?

• A. Append instructions telling Claude to "only report findings you are highly confident are genuine problems."
• B. Add explicit criteria defining which issues to report, such as bugs and security defects, and which issues to skip, such as minor style preferences and accepted local patterns.
• C. Implement a secondary classification model that filters Claude's findings according to predicted developer acceptance.
• D. Ask Claude to rate every finding's confidence from 1 to 10 and include only findings rated 8 or higher.

Your control_device tool manages smart home devices through external APIs. When a device doesn't respond within the timeout period, the tool returns an error. Production logs show that the agent simply tells users "the device is not responding" without offering helpful next steps. Which error response structure would best enable the agent to provide useful follow-up?

• A. Set is_error: true with a message explaining the likely cause and suggesting troubleshooting steps the agent can offer the user.
• B. Set is_error: true with a brief "Device offline" message and provide a separate tool the agent can call to retrieve context-specific troubleshooting suggestions.
• C. Set is_error: false with an optimistic message indicating the command was dispatched successfully but device acknowledgment is still pending.
• D. Set is_error: true with a structured technical error containing the device ID, timeout duration, and raw API response code for debugging purposes.