Dr. QuantMaster MCP Server

Instructions

Implementation Reference

• src/tools/index.ts:1760-1769 (handler)

  The handler function for 'interpret_test' tool. It takes input arguments like test_name, p_value, statistic, and returns a structured interpretation, primarily based on whether p-value < 0.05, advising to report effect sizes beyond p-values.

  function handleInterpretTest(args: Record<string, unknown>) {
    const pValue = args.p_value as number;
    return {
      test_name: args.test_name,
      p_value: pValue,
      statistic: args.statistic,
      conclusion: pValue < 0.05 ? "통계적으로 유의함 (p < .05)" : "통계적으로 유의하지 않음 (p ≥ .05)",
      effect_size_note: "p-value 외에 효과크기도 보고 필요"
    };
  }

• src/tools/index.ts:508-520 (schema)

  The input schema definition for the 'interpret_test' tool within the exported tools array, specifying properties like test_name, statistic, p_value (required), df, context.

  name: "interpret_test",
  description: "통계 검정 결과 해석",
  inputSchema: {
    type: "object",
    properties: {
      test_name: { type: "string", description: "검정명" },
      statistic: { type: "number", description: "검정통계량" },
      p_value: { type: "number", description: "p-value" },
      df: { type: "number", description: "자유도" },
      context: { type: "string", description: "맥락" },
    },
    required: ["test_name", "p_value"],
  },

• src/tools/index.ts:856-857 (registration)

  Registration of the 'interpret_test' tool in the handleToolCall switch statement, mapping the tool name to its handler function handleInterpretTest.

  case "interpret_test":
    return handleInterpretTest(args);

• src/tools/index.ts:508-520 (registration)

  The tool object in the exported 'tools' array, registering the name, description, and schema for the MCP tool system.

  name: "interpret_test",
  description: "통계 검정 결과 해석",
  inputSchema: {
    type: "object",
    properties: {
      test_name: { type: "string", description: "검정명" },
      statistic: { type: "number", description: "검정통계량" },
      p_value: { type: "number", description: "p-value" },
      df: { type: "number", description: "자유도" },
      context: { type: "string", description: "맥락" },
    },
    required: ["test_name", "p_value"],
  },