メニュー
Use this skill when working with Sahaidachny task artifacts. Activated when: - Working with task planning documents - Reading/updating user stories, design decisions, implementation plans - Any task folder (task-XX/) operations
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use this skill when working with Sahaidachny task planning. Activated when: - User mentions "task", "planning", "user stories", "design decisions" - Working in a `task-XX/` directory - Any `/saha:*` command is invoked
Comprehensive test quality and completeness analysis across 6 dimensions. Use when: - Verifying ALL needed tests exist (cross-ref user stories, code changes, test specs) - Reviewing test implementations before QA - Verifying tests actually test real behavior - Detecting over-mocking, poor assertions, missing edge cases - Identifying flaky patterns and test smells
Guidance for measuring cognitive complexity with complexipy.
Guidance for running Ruff and interpreting lint results.
Guidance for evaluating test quality and detecting hollow tests.
Guidance for running the ty type checker.
| name | task-structure |
| description | Use this skill when working with Sahaidachny task artifacts. Activated when: - Working with task planning documents - Reading/updating user stories, design decisions, implementation plans - Any task folder (task-XX/) operations |
| version | 0.1.0 |
Sahaidachny uses a hierarchical task structure for planning complex features before implementation.
task-XX-short-name/
├── README.md # Task status dashboard
├── task-description.md # Architecture + technical overview
├── user-stories/
│ ├── US-001-feature-name.md # Detailed requirements
│ └── ...
├── design-decisions/
│ ├── README.md # Index of decisions
│ ├── DD-001-decision-name.md # Architecture decisions
│ └── ...
├── code-changes/
│ └── ...
├── implementation-plan/
│ ├── README.md # Phase status tracking
│ ├── phase-01-name.md # Execution phases with steps
│ └── ...
├── test-specs/
│ ├── e2e/ # End-to-end test specs
│ ├── integration/ # Integration test specs
│ └── unit/ # Unit test specs
└── research/
└── ...
# US-XXX: [Title]
**Status:** Draft | Review | Approved | Done
**Priority:** Critical | High | Medium | Low
## User Story
As a [role], I want [feature] so that [benefit].
## Acceptance Criteria
- [ ] AC1: [Criterion]
- [ ] AC2: [Criterion]
## Edge Cases
- Edge case 1
# Phase XX: [Name]
**Status:** Pending | In Progress | Completed
## Steps
### Step 1: [Name]
- [ ] Action item 1
- [ ] Action item 2
## Success Criteria
- [ ] Criterion 1
| Marker | Meaning |
|---|---|
[ ] | Pending / Not started |
[~] | In progress |
[x] | Completed |
| Artifact | Pattern | Example |
|---|---|---|
| Task folder | task-XX-short-name/ | task-05-user-auth/ |
| User Story | US-XXX-short-name.md | US-001-login-form.md |
| Design Decision | DD-XXX-short-name.md | DD-001-jwt-vs-session.md |
| Phase | phase-XX-short-name.md | phase-01-setup.md |