Phase 1b: Rename all ~94 commands across 12 plugins to /<noun> <action> sub-command pattern. Git-flow consolidated from 8→5 commands (commit variants absorbed into --push/--merge/--sync flags). Dispatch files, name: frontmatter, and cross-reference updates for all plugins. Phase 2: Design documents for 8 new plugins in docs/designs/. Phase 3: Scaffold 8 new plugins — saas-api-platform, saas-db-migrate, saas-react-platform, saas-test-pilot, data-seed, ops-release-manager, ops-deploy-pipeline, debug-mcp. Each with plugin.json, commands, agents, skills, README, and claude-md-integration. Marketplace grows from 12→20. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2.4 KiB
2.4 KiB
name, description
| name | description |
|---|---|
| test generate | Generate test cases for functions, classes, or modules with appropriate patterns |
/test generate
Generate comprehensive test cases for specified code targets.
Visual Output
+----------------------------------------------------------------------+
| TEST-PILOT - Generate Tests |
+----------------------------------------------------------------------+
Usage
/test generate <target> [--type=unit|integration] [--style=aaa|bdd]
Target: File path, class name, function name, or module path Type: Test type — defaults to unit Style: Test style — defaults to arrange-act-assert (aaa)
Skills to Load
- skills/test-patterns.md
- skills/mock-patterns.md
- skills/framework-detection.md
Process
-
Analyze Target
- Read the target source code
- Identify public functions, methods, and classes
- Map input types, return types, and exceptions
- Detect dependencies that need mocking
-
Determine Test Strategy
- Pure functions: direct input/output tests
- Functions with side effects: mock external calls
- Class methods: test through public interface
- Integration points: setup/teardown with real or fake dependencies
-
Generate Test Cases
- Happy path: standard inputs produce expected outputs
- Edge cases: empty inputs, None/null, boundary values
- Error paths: invalid inputs, exceptions, error conditions
- Type variations: different valid types if applicable
-
Write Test File
- Follow project conventions for test file location
- Use detected framework syntax (pytest/Jest/Vitest)
- Include docstrings explaining each test case
- Group related tests in classes or describe blocks
-
Verify
- Check test file compiles/parses
- Verify imports are correct
- Confirm mock targets match actual module paths
Output Format
## Generated Tests for `module.function_name`
### Test File: tests/unit/test_module.py
### Test Cases (7 total)
1. test_function_returns_expected_for_valid_input
2. test_function_handles_empty_input
3. test_function_raises_on_invalid_type
4. test_function_boundary_values
5. test_function_none_input
6. test_function_large_input
7. test_function_concurrent_calls (if applicable)
### Dependencies Mocked
- database.connection (unittest.mock.patch)
- external_api.client (fixture)