You are integrating Claude Code into your Continuous Integration/Continuous Deployment (CI/CD) pipeline. The system runs automated code reviews, generates test cases, and provides feedback on pull requests. You need to design prompts that provide actionable feedback and minimize false positives.
After deploying automated code review, developers report that approximately 35% of findings are false positives following consistent patterns: style suggestions that contradict team conventions, security warnings for patterns that are safe in the deployment environment, and performance suggestions that would degrade this particular use case.
You want to reduce false positives while enabling the model to generalize its judgment to novel code patterns it has not seen before.
Which approach is most effective?

• A. Create a comprehensive specification of every pattern that must not be flagged and include the complete document in the system prompt.
• B. Use keyword-based post-processing to remove findings containing terms such as "convention,"
  "context-dependent," or "trade-off."
• C. Add general instructions telling Claude to be conservative and report only definite issues.
• D. Include few-shot examples containing annotated code snippets that distinguish acceptable project patterns from genuine issues in each category.

A customer returns 4 hours after the initial session about the same billing dispute. The previous
32-turn session contains lookup_order results showing "Status: PENDING, Expected resolution:
24-48 hours." In testing, you observe that when resuming sessions with stale tool results, the agent often references the outdated data in responses (e.g., "I see your refund is still being processed") even after subsequent fresh tool calls return different information. What approach most reliably handles returning customers?

• A. Start a new session, inject a structured summary of the previous interaction (issue type, actions taken, resolution status), then make fresh tool calls before engaging.
• B. Resume with full history but filter out previous tool_result messages before resuming, keeping only the human/assistant turns so the agent must re-fetch needed data.
• C. Resume with full history and add a system prompt instruction telling the agent to always prefer the most recent tool results when multiple calls to the same tool exist in context.
• D. Resume with full history and configure the agent to automatically re-call all previously-used tools at session start to ensure data freshness.

The coordinator agent has AgentDefinition objects configured for all four specialized subagents, each with appropriate descriptions, prompts, and tool restrictions. During testing, you notice that the coordinator correctly reasons about when to delegate--it generates messages such as, "I'll ask the web-search agent to find sources on this topic"--but no subagent execution occurs. The coordinator then proceeds as if the delegation happened and continues with incomplete information. Logs show no errors. What is the most likely cause?

• A. The coordinator's allowedTools configuration does not include "Agent"--called "Task" in older SDK releases--so it cannot invoke the tool required to spawn subagents.
• B. Subagent context isolation prevents task descriptions from reaching subagents unless explicit context forwarding is configured in ClaudeAgentOptions.
• C. The AgentDefinition objects are configured correctly, but the coordinator's system prompt does not explicitly list the available subagent types.
• D. The coordinator's max_tokens setting is too low, causing the subagent invocation to be truncated before the agent-type parameter is specified.

Your automated review calls the Claude API for each pull request, using tool_use with a report_findings tool that returns a JSON array of finding objects. Each object contains file_path, line_number, severity, category, and description. During testing on a large pull request touching more than 30 files, the response reaches the max_tokens limit and is truncated in the middle of the JSON, causing your pipeline's parser to fail. What is the most effective way to handle this?

• A. Switch from tool_use to prompting Claude to return findings as a Markdown list.
• B. Split the review into multiple API calls that each analyze a subset of the changed files, and then merge the resulting findings arrays.
• C. Increase max_tokens to the model's maximum and instruct Claude to keep each finding description under 50 words.
• D. Add retry logic that detects truncated JSON and resends the request with instructions to report only critical-and high-severity findings.