QA-MCP: Test Standardization & Orchestration Server

Instructions

Implementation Reference

• src/qa_mcp/tools/compose.py:19-106 (handler)

  Main handler function that implements the core logic of the suite.compose tool, including test case selection, coverage analysis, and suite composition.
  def compose_suite( testcases: list[dict], target: str, sprint: str | None = None, max_duration_minutes: int | None = None, custom_criteria: dict[str, Any] | None = None, ) -> dict: """ Compose a test suite from a collection of test cases. Args: testcases: List of test cases in QA-MCP standard format target: Suite type - 'smoke', 'sanity', 'regression', 'e2e' sprint: Sprint name/number for context max_duration_minutes: Maximum suite duration (overrides default) custom_criteria: Custom selection criteria Returns: Dictionary containing: - suite: Composed suite with selected test cases - selection_rationale: Why each test was included/excluded - coverage_summary: What's covered and what's not - recommendations: Suggestions for improving coverage """ # Parse test cases parsed_cases = [] parse_errors = [] for idx, tc_dict in enumerate(testcases): try: tc = TestCase(**tc_dict) parsed_cases.append(tc) except Exception as e: parse_errors.append(f"Test case {idx}: {str(e)}") if parse_errors: return { "suite": None, "selection_rationale": [], "coverage_summary": {}, "recommendations": [], "errors": parse_errors, } # Get suite rules suite_type = SuiteType(target.lower()) rules = SUITE_RULES.get(target.lower(), SUITE_RULES["regression"]) # Override max duration if provided if max_duration_minutes: rules = {**rules, "max_duration_minutes": max_duration_minutes} # Select test cases selected, excluded, rationale = _select_testcases(parsed_cases, suite_type, rules) # Build coverage summary coverage_summary = _build_coverage_summary(parsed_cases, selected, suite_type) # Generate recommendations recommendations = _generate_recommendations(parsed_cases, selected, excluded, suite_type) # Calculate total duration total_duration = sum( tc.estimated_duration_minutes or 5 # Default 5 min per test for tc in selected ) # Build suite composition suite = SuiteComposition( suite_type=suite_type, name=f"{target.upper()} Suite" + (f" - {sprint}" if sprint else ""), description=rules.get("description", f"{target} test suite"), testcases=[tc.id or f"TC-{idx}" for idx, tc in enumerate(selected)], total_duration_minutes=total_duration, coverage_summary=coverage_summary, rationale=f"{len(selected)} test case seçildi, toplam {total_duration} dakika", ) return { "suite": suite.model_dump(), "selected_testcases": [tc.model_dump() for tc in selected], "excluded_count": len(excluded), "selection_rationale": rationale, "coverage_summary": coverage_summary, "recommendations": recommendations, "duration_warning": total_duration > rules.get("max_duration_minutes", 999), }
• src/qa_mcp/server.py:238-265 (registration)

  Registration of the suite.compose tool in the MCP server, including name, description, and input schema.
  Tool( name="suite.compose", description="Test case listesinden Smoke/Regression/E2E suite oluşturur", inputSchema={ "type": "object", "properties": { "testcases": { "type": "array", "items": {"type": "object"}, "description": "Test case listesi", }, "target": { "type": "string", "enum": ["smoke", "sanity", "regression", "e2e"], "description": "Suite tipi", }, "sprint": { "type": "string", "description": "Sprint adı/numarası (opsiyonel)", }, "max_duration_minutes": { "type": "integer", "description": "Maksimum suite süresi (dakika)", }, }, "required": ["testcases", "target"], }, ),
• src/qa_mcp/server.py:376-384 (handler)

  Tool dispatcher in call_tool() that invokes the compose_suite handler for suite.compose calls.
  elif name == "suite.compose": result = compose_suite( testcases=arguments["testcases"], target=arguments["target"], sprint=arguments.get("sprint"), max_duration_minutes=arguments.get("max_duration_minutes"), ) audit_log(name, arguments, f"Composed {arguments['target']} suite")