analyze_test_cases_duplicates_semantic
Identify duplicate test cases using semantic analysis and LLM-powered clustering to optimize test suites and reduce redundancy in QA workflows.
Instructions
🧠 Advanced semantic duplicate analysis using LLM-powered step clustering and two-phase analysis
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| project_key | Yes | Project key (e.g., 'ANDROID', 'IOS') | |
| suite_id | No | Optional: Analyze specific test suite ID | |
| test_case_keys | No | Optional: Analyze specific test case keys instead of suite | |
| similarity_threshold | No | Test case similarity threshold percentage (50-100, default: 80) | |
| step_clustering_threshold | No | Step clustering threshold percentage (50-100, default: 85) | |
| analysis_mode | No | Analysis mode: basic (fast), semantic (LLM-powered), hybrid (both) | hybrid |
| use_step_clustering | No | Enable two-phase clustering (step clusters first, then test case clusters) | |
| use_medoid_selection | No | Use medoid-based representative selection instead of heuristic | |
| include_semantic_insights | No | Generate semantic insights about workflows and patterns | |
| format | No | Output format | markdown |
| include_similarity_matrix | No | Include detailed similarity matrix in output | |
| include_clickable_links | No | Include clickable links to Zebrunner web UI (markdown format only) |