先試後買

A. Add routing instructions to the system prompt specifying that dependency-related questions should use MCP tools rather than Grep.

B. Split analyze_dependencies into granular tools ( list_imports , resolve_transitive_deps , detect_circular_deps ) so each has a focused purpose less likely to overlap with Grep.

C. Expand MCP tool descriptions to detail capabilities and outputs-e.g., "Builds dependency graph showing direct imports, transitive dependencies, and cycles."

D. Remove Grep from available tools when the MCP server is connected to eliminate functional overlap.

A. Add the venue server to ~/.claude.json and the playlist server to .mcp.json .

B. Add both servers to your local ~/.claude.json .

C. Add both servers to the project-level .mcp.json file.

D. Add the venue server to .mcp.json and the playlist server to ~/.claude.json .

A. Implement sentiment analysis that monitors for frustration indicators (negative language, repeated questions, exclamation marks) and triggers escalation when the frustration score exceeds a configured threshold.

B. Build a rules engine that maps specific issue types, customer segments, and product categories to escalation decisions, removing the need for model judgment calls.

C. Configure the agent to escalate after three consecutive tool calls that fail to resolve the customer's stated issue, ensuring a reasonable attempt before involving a human.

D. Instruct the agent to escalate when the customer requests a human, when the issue requires policy exceptions, or when the agent cannot make meaningful progress.

A. Define a fixed sequence of investigation steps upfront-grep for error patterns, then read error handlers, then check database queries, then examine middleware-executing each step regardless of intermediate findings.

B. Run parallel worker agents that simultaneously investigate all four layers, then synthesize their findings to identify where the error originates.

C. Have the agent dynamically generate investigation subtasks based on what it discovers at each step, adapting its exploration plan as new information about the error path emerges.

D. Have the agent first create a comprehensive plan mapping all code paths through the endpoint before beginning any file exploration or code reading.

A. Implement a hook to intercept tool calls, when the refund process amount exceeds $500, block it and invoke human escalation.

B. Modify the refund tool to return an error with message "Amount exceeds policy limit-please escalate" when the threshold is exceeded.

C. Add few-shot examples to the prompt showing correct escalation behavior at various refund amounts ($400, $500, $600).

D. Strengthen the system prompt with emphatic language: "CRITICAL POLICY: Refunds over $500 MUST trigger human escalation. NEVER process these directly."

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 and configure the agent to automatically re-call all previously used tools at session start to ensure data freshness.

C. 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.

D. 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.

A. Routing-classify each PR by type (feature, bugfix, refactor) first, then route to different review prompts optimized for that category.

B. Prompt chaining-break the review into sequential steps where each aspect (style, security, documentation) is analyzed separately, with outputs combined in a final synthesis step.

C. Orchestrator-workers-have a central LLM analyze each PR to dynamically determine which checks are needed, then delegate to specialized worker LLMs for each identified subtask.

D. Single comprehensive prompt-include all three instructions in one prompt and let the model handle all three aspects simultaneously.