원클릭으로
std-orchestrator
Orchestrate STP → STD pipeline (generates comprehensive STD YAML only)
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Orchestrate STP → STD pipeline (generates comprehensive STD YAML only)
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Classify PR review comments on STP/STD documents into auto-fixable vs needs-human categories. Maps free-text feedback to QualityFlow domain rules for automated fix routing.
Resolve Jira ID to project configuration and load project context
Generate comprehensive v2.1-ENHANCED STD YAML with pattern metadata, variables, test structure from ALL STP scenarios (single file)
Generate test stubs with PSE docstrings from STD YAML — language and framework driven by project config
Detect test conventions from a codebase to drive test generation without tier config
Discover code entry points from Jira data when no PR data exists
| name | std-orchestrator |
| description | Orchestrate STP → STD pipeline (generates comprehensive STD YAML only) |
| model | claude-opus-4-6 |
| phase_support | true |
| default_phase | phase1 |
Coordinates the Software Test Description (STD) generation workflow by:
Output:
stp_file_path: Path to the STP markdown file (e.g., outputs/stp/PROJ-66855/PROJ-66855_test_plan.md)jira_id: The Jira ticket ID (e.g., "PROJ-66855")output_dir: Base directory for outputs (defaults to outputs/std/{JIRA_ID}/)phase: phase1 (default) or phase2Phase 1 (default):
phase=phase1pass body__test__ = False for Python, PendingIt() for Go)Phase 2:
phase=phase2Execute the following steps in order:
Read the STP file and extract all test scenarios from Section III (Test Scenarios & Traceability).
Expected bullet-based format:
- **[PROJ-12345]** — As a user, I want to reset a VM
- *Test Scenario:* [Tier 1] Verify basic reset operation succeeds
- *Priority:* P0
- **[PROJ-12345]** — As a user, I want to verify reset preserves state
- *Test Scenario:* [Tier 1] Verify reset preserves pod UID
- *Priority:* P0
Parse and extract:
**[ID]**)—)[Tier 1], [Tier 2][unit], [functional], [integration], [e2e]*Test Scenario:* line)[EXISTING_COVERAGE], [PARTIAL_COVERAGE]Store as:
scenarios:
- scenario_id: 1
tier: "Tier 1" # tier mode
test_type: "functional" # auto mode (one or the other)
priority: "P0"
description: "Verify basic reset operation succeeds"
coverage_status: "NEW" # optional, defaults to NEW
- scenario_id: 2
tier: "Tier 1"
priority: "P0"
description: "Verify reset preserves pod UID"
coverage_status: "EXISTING_COVERAGE"
covered_by:
- test_function: "TestResetPreservesPodUID"
test_file: "pkg/compute/reset_test.go"
Generate ONE comprehensive STD file for ALL scenarios:
1.5. Extract Source Constants (if present):
Check if the STP contains a ## Source Constants or #### Section III.2 — Source Constants section
If found, parse the markdown table into a structured array:
source_constants:
- name: "SENTINEL"
value: "# --- managed section - do not edit ---"
source_file: "pkg/scripts/sync.sh"
line: 14
Values are in backtick-wrapped cells — strip the backticks but preserve the exact content
If the section is not present, set source_constants to an empty array
Call std-generator skill ONCE with:
source_constants array (from Step 1.5, may be empty)Output file:
outputs/std/{JIRA_ID}/{JIRA_ID}_test_description.yamloutputs/std/PROJ-66855/PROJ-66855_test_description.yamlValidate STD output:
After STD YAML is validated, call appropriate code generators.
Routing depends on test_strategy_mode in the STD YAML document_metadata:
code_generation_config.language from STD YAML| Language | Generator | Output |
|---|---|---|
go | stub-generator / test-generator | Go test stubs/implementations |
python | stub-generator / test-generator | Python test stubs/implementations |
coverage_status: NEW or PARTIAL_COVERAGEEXISTING_COVERAGE scenarioscode_generation_config so the generator uses detected framework conventionsOutput files:
outputs/std/{JIRA_ID}/
├── go-tests/ (if language is Go, or Tier 1 in tier mode)
│ └── *_stubs_test.go (Phase 1: stubs, Phase 2: implementation)
└── python-tests/ (if language is Python, or Tier 2 in tier mode)
└── test_*_stubs.py (Phase 1: stubs, Phase 2: implementation)
Create a summary report with:
Output format:
---
status: success
component: std-orchestrator
jira_id: PROJ-66855
phase: phase1 # or phase2
stp_file: outputs/stp/PROJ-66855/PROJ-66855_test_plan.md
output_dir: outputs/std/PROJ-66855/
execution_summary:
total_stp_scenarios: 12
tier_1_scenarios: 9
tier_2_scenarios: 3
std_file_generated: "PROJ-66855_test_description.yaml"
scenarios_in_std: 12
total_duration: "2 minutes"
code_generation:
phase: phase1
go_tests:
file_count: 2
test_count: 9
status: "stubs_generated" # or "implementation_generated" for phase2
python_tests:
file_count: 1
test_count: 3
status: "stubs_generated"
validation_results:
std_file:
file: PROJ-66855_test_description.yaml
status: valid
yaml_syntax: passed
required_sections: passed
scenarios_count: 12
errors: []
warnings: []
notes:
- "STD YAML generated as internal format"
- "Use /generate-tests for implementations"
---
Simple structure (STD YAML only):
outputs/std/PROJ-66855/
├── PROJ-66855_test_description.yaml (NEW - comprehensive STD for ALL scenarios)
└── std_generation_summary.yaml (summary report)
Key design:
This orchestrator calls 1 specialized skill:
Architecture:
If STP parsing fails:
If std-generator fails for a scenario:
Set overall status to:
success: All scenarios processed successfullypartial: Some scenarios failed, but >50% succeedederror: >50% scenarios failed or critical errorThe orchestration is complete when:
Minimum acceptable outcome:
Before marking orchestration as complete, validate:
User command (Phase 1 - default):
Generate STD/PSE/Code for PROJ-66855
Orchestrator execution (Phase 1):
1. Read outputs/stp/PROJ-66855/PROJ-66855_test_plan.md
2. Parse Section III → 12 scenarios found (9 Tier 1, 3 Tier 2)
3. Call std-generator ONCE → PROJ-66855_test_description.yaml
4. Validate STD YAML
5. Call stub-generator → 9 Go stubs + 3 Python stubs
7. Generate summary → std_generation_summary.yaml
8. Report to user: "✅ Generated Phase 1 test stubs for 12 scenarios"
Example output (Phase 1):
✅ Phase 1 Test Stubs Generated!
📄 Input: outputs/stp/PROJ-66855/PROJ-66855_test_plan.md
📊 Summary:
- STP scenarios: 12 (9 Tier 1, 3 Tier 2)
- STD file: PROJ-66855_test_description.yaml (internal format)
- Phase: 1 (Design stubs with PSE docstrings)
📁 Output:
- outputs/std/PROJ-66855/PROJ-66855_test_description.yaml
- outputs/std/PROJ-66855/go-tests/ (9 test stubs with PSE comments)
- outputs/std/PROJ-66855/python-tests/ (3 test stubs with PSE docstrings)
📋 Phase 1 Checklist:
- [ ] STP link in module docstring
- [ ] Tests grouped in class with shared preconditions
- [ ] Each test has: Preconditions, Steps, Expected
- [ ] Each test verifies ONE thing with ONE Expected
- [ ] Test bodies contain only 'pass'
✅ Ready for design review!
📌 Next steps:
1. Review the test stubs
2. Submit PR for design review
3. After approval, run:
- /generate-tests PROJ-66855 (test implementations)
End of STD Orchestrator Skill