QA-MCP: Test Standardization & Orchestration Server

Instructions

Implementation Reference

• src/qa_mcp/tools/normalize.py:20-82 (handler)

  Core handler function implementing the testcase.normalize tool logic. Detects input format automatically if needed, parses the input_data accordingly (gherkin, markdown, json, plain), fills missing fields, normalizes values, and returns the standardized test case with metadata.

  def normalize_testcase(
      input_data: str | dict,
      source_format: str = "auto",
  ) -> dict:
      """
      Normalize a test case from various formats to QA-MCP standard.
  
      Args:
          input_data: Test case in string (markdown/gherkin) or dict format
          source_format: Source format - 'auto', 'markdown', 'gherkin', 'json', 'plain'
  
      Returns:
          Dictionary containing:
          - testcase: Normalized test case in standard format
          - source_format_detected: Detected source format
          - transformations: List of transformations applied
          - warnings: Any warnings during normalization
      """
      transformations = []
      warnings = []
  
      # Detect format if auto
      if source_format == "auto":
          source_format = _detect_format(input_data)
          transformations.append(f"Format otomatik tespit edildi: {source_format}")
  
      # Parse based on format
      try:
          if source_format == "gherkin":
              testcase, parse_warnings = _parse_gherkin(input_data)
          elif source_format == "markdown":
              testcase, parse_warnings = _parse_markdown(input_data)
          elif source_format == "json" or isinstance(input_data, dict):
              testcase, parse_warnings = _parse_json(input_data)
          else:  # plain text
              testcase, parse_warnings = _parse_plain_text(input_data)
  
          warnings.extend(parse_warnings)
  
      except Exception as e:
          return {
              "testcase": None,
              "source_format_detected": source_format,
              "transformations": transformations,
              "warnings": [f"Parse hatası: {str(e)}"],
              "error": str(e),
          }
  
      # Validate and fill missing fields
      testcase, fill_warnings = _fill_missing_fields(testcase)
      warnings.extend(fill_warnings)
      transformations.append("Eksik alanlar varsayılan değerlerle dolduruldu")
  
      # Normalize values
      testcase = _normalize_values(testcase)
      transformations.append("Değerler standartlaştırıldı")
  
      return {
          "testcase": testcase.model_dump(),
          "source_format_detected": source_format,
          "transformations": transformations,
          "warnings": warnings,
      }

• src/qa_mcp/server.py:165-182 (registration)

  Registration of the 'testcase.normalize' tool in the MCP server, including name, description, and input schema.

  Tool(
      name="testcase.normalize",
      description="Farklı formatlardaki test case'leri QA-MCP standardına çevirir",
      inputSchema={
          "type": "object",
          "properties": {
              "input_data": {
                  "type": ["string", "object"],
                  "description": "Normalize edilecek test case (markdown, gherkin, json veya plain text)",
              },
              "source_format": {
                  "type": "string",
                  "enum": ["auto", "markdown", "gherkin", "json", "plain"],
                  "description": "Kaynak format (default: auto)",
              },
          },
          "required": ["input_data"],
      },

• src/qa_mcp/server.py:168-181 (schema)

  Input schema defining parameters for testcase.normalize: input_data (string or object, required) and optional source_format.

  inputSchema={
      "type": "object",
      "properties": {
          "input_data": {
              "type": ["string", "object"],
              "description": "Normalize edilecek test case (markdown, gherkin, json veya plain text)",
          },
          "source_format": {
              "type": "string",
              "enum": ["auto", "markdown", "gherkin", "json", "plain"],
              "description": "Kaynak format (default: auto)",
          },
      },
      "required": ["input_data"],

• src/qa_mcp/server.py:343-352 (handler)

  Dispatch handler in server.call_tool that invokes the normalize_testcase function with parsed arguments and logs audit.

  elif name == "testcase.normalize":
      result = normalize_testcase(
          input_data=arguments["input_data"],
          source_format=arguments.get("source_format", "auto"),
      )
      audit_log(
          name,
          arguments,
          f"Normalized from {result.get('source_format_detected', 'unknown')}",
      )