メニュー
AI agent creates structured implementation plans with task breakdown, dependency mapping, risk assessment, and file-level detail. Use when planning features, projects, or complex implementations.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
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。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/doanchienthangdev/omgkit --skill writing-plansコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Automatic design system context injection for UI consistency
AI agent practices test-first development with the Red-Green-Refactor cycle for confident, well-designed code. Use when implementing features, fixing bugs, or establishing testing practices.
The agent enforces mandatory test completion before any task or feature can be marked as done, ensuring code quality through strict validation gates and evidence-based completion criteria.
SOC 職業分類に基づく
SKILL.md を表示中
| name | writing-plans |
| description | AI agent creates structured implementation plans with task breakdown, dependency mapping, risk assessment, and file-level detail. Use when planning features, projects, or complex implementations. |
| Feature | Description | Guide |
|---|---|---|
| Plan Structure | Comprehensive template | Summary, design, tasks, testing, risks |
| Task Breakdown | Manageable work units | 2-4 hours, single owner, clear criteria |
| Dependency Mapping | Sequence and parallelize | Critical path, hard/soft dependencies |
| Risk Assessment | Anticipate problems | Likelihood x Impact matrix |
| Effort Estimation | Realistic timing | Three-point: (O + 4M + P) / 6 |
| File-Level Detail | Exact code locations | New files, modifications, line numbers |
# Plan Structure
# Plan: [Feature Name]
## Metadata
- Status: Draft | Approved | In Progress
- Estimated Duration: [X days]
- Priority: P0 | P1 | P2
## Goals & Non-Goals
Goals: [What this WILL accomplish]
Non-Goals: [What this will NOT address]
## Design Overview
[High-level approach, architecture diagram]
## Implementation Tasks
### Phase 1: [Foundation] - [X days]
#### Task 1.1: [Name]
- Description: [What]
- Estimate: [X hours]
- Files: `src/file.ts:45-60`
- Dependencies: None
- Acceptance Criteria:
- [ ] [Criterion 1]
- [ ] [Criterion 2]
## Risks & Mitigations
| Risk | Likelihood | Impact | Mitigation |
|------|------------|--------|------------|
| [Risk] | High | High | [Strategy] |
# Dependency Diagram
[START]
|
+----------+----------+
| | |
[Task 1] [Task 2] [Task 3] <- Parallel
| | |
+-----+----+ |
| |
[Task 4] [Task 5]
| |
+-------+-------+
|
[Task 6] <- Integration
|
[COMPLETE]
Critical Path: 1 -> 4 -> 6 (determines min duration)
# Effort Estimation
Three-Point: (Optimistic + 4*MostLikely + Pessimistic) / 6
Example - OAuth Implementation:
- Optimistic: 4 hours (works first try)
- Most Likely: 8 hours (normal dev)
- Pessimistic: 16 hours (issues)
Estimate = (4 + 32 + 16) / 6 = 8.7 hours
Buffers:
- Individual tasks: 20%
- Phase total: 30%
- Project total: 40%
| Do | Avoid |
|---|---|
| Keep tasks to 2-4 hour chunks | Vague tasks ("implement feature") |
| Include exact file locations | Skipping background/context |
| Specify clear acceptance criteria | Ignoring non-functional requirements |
| Map dependencies explicitly | Underestimating integration work |
| Build in buffer for unknowns | Planning more than 2 weeks in detail |
| Include testing in the plan | Hiding risks or uncertainties |
| Review plan with stakeholders | Assuming requirements are complete |
| Update plan as you learn | Forgetting rollback plans |
executing-plans - Follow plans systematicallythinking-sequentially - Structure reasoning for plansbrainstorming-ideas - Generate options for plansverifying-before-completion - Validate plan completion