feature-review
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseTable of Contents
目录
Verification
验证
Run to verify scoring logic after changes.
make test-feature-review- 4. Upload to GitHub
- Workflow
- Phase 1: Feature Discovery ()
feature-review:inventory-complete - Phase 2: Classification ()
feature-review:classified - Phase 3: Scoring ()
feature-review:scored - Phase 4: Tradeoff Analysis ()
feature-review:tradeoffs-analyzed - Phase 5: Gap Analysis & Suggestions ()
feature-review:suggestions-generated - Phase 6: GitHub Integration ()
feature-review:issues-created - Configuration
- Configuration File
- Guardrails
- Required TodoWrite Items
- Integration Points
- Output Format
- Feature Inventory Table
- Suggestion Report
- Feature Suggestions
- High Priority (Score > 2.5)
- Related Skills
- Reference
修改后运行来验证评分逻辑。
make test-feature-review- 4. 上传至GitHub
- 工作流程
- 阶段1：功能发现（）
feature-review:inventory-complete - 阶段2：分类（）
feature-review:classified - 阶段3：评分（）
feature-review:scored - 阶段4：权衡分析（）
feature-review:tradeoffs-analyzed - 阶段5：差距分析与建议（）
feature-review:suggestions-generated - 阶段6：GitHub集成（）
feature-review:issues-created - 配置
- 配置文件
- 防护规则
- 必填TodoWrite项
- 集成点
- 输出格式
- 功能盘点表格
- 建议报告
- 功能建议
- 高优先级（评分>2.5）
- 相关技能
- 参考资料
Feature Review
功能评审
Review implemented features and suggest new ones using evidence-based prioritization. Create GitHub issues for accepted suggestions.
基于循证优先级排序方法评审已实现功能并提出新功能建议。为已采纳的建议创建GitHub Issues。
Philosophy
设计理念
Feature decisions rely on data. Every feature involves tradeoffs that require evaluation. This skill uses hybrid RICE+WSJF scoring with Kano classification to prioritize work and generates actionable GitHub issues for accepted suggestions.
功能决策需基于数据。每个功能都涉及需要评估的权衡。本技能采用RICE+WSJF混合评分法结合Kano分类来确定工作优先级，并为已采纳的建议生成可执行的GitHub Issues。
When To Use
适用场景
- Roadmap reviews (sprint planning, quarterly reviews).
- Retrospective evaluations.
- Planning new development cycles.
- 路线图评审（迭代规划、季度评审）。
- 回顾性评估。
- 规划新的开发周期。
When NOT To Use
不适用场景
- Emergency bug fixes.
- Simple documentation updates.
- Active implementation (use ).
scope-guard
- 紧急Bug修复。
- 简单文档更新。
- 活跃开发中的功能（请使用）。
scope-guard
Quick Start
快速开始
1. Inventory Current Features
1. 盘点现有功能
Discover and categorize existing features:
bash
/feature-review --inventory发现并分类现有功能：
bash
/feature-review --inventory2. Score and Classify
2. 评分与分类
Evaluate features against the prioritization framework:
bash
/feature-review基于优先级框架评估功能：
bash
/feature-review3. Generate Suggestions
3. 生成建议
Review gaps and suggest new features:
bash
/feature-review --suggest评审差距并提出新功能建议：
bash
/feature-review --suggest4. Upload to GitHub
4. 上传至GitHub
Create issues for accepted suggestions:
bash
/feature-review --suggest --create-issues为已采纳的建议创建Issues：
bash
/feature-review --suggest --create-issuesWorkflow
工作流程
Phase 1: Feature Discovery (feature-review:inventory-complete
)
feature-review:inventory-complete阶段1：功能发现（feature-review:inventory-complete
）
feature-review:inventory-completeIdentify features by analyzing:
- Code artifacts: Entry points, public APIs, and configuration surfaces.
- Documentation: README lists, CHANGELOG entries, and user docs.
- Git history: Recent feature commits and branches.
Output: Feature inventory table.
通过分析以下内容识别功能：
- 代码工件：入口点、公开API和配置界面。
- 文档：README列表、CHANGELOG条目和用户文档。
- Git历史：近期功能提交和分支。
输出： 功能盘点表格。
Phase 2: Classification (feature-review:classified
)
feature-review:classified阶段2：分类（feature-review:classified
）
feature-review:classifiedClassify each feature along two axes:
Axis 1: Proactive vs Reactive
| Type | Definition | Examples |
|---|---|---|
| Proactive | Anticipates user needs. | Suggestions, prefetching. |
| Reactive | Responds to explicit input. | Form handling, click actions. |
Axis 2: Static vs Dynamic
| Type | Update Pattern | Storage Model |
|---|---|---|
| Static | Incremental, versioned. | File-based, cached. |
| Dynamic | Continuous, streaming. | Database, real-time. |
See classification-system.md for details.
从两个维度对每个功能进行分类：
维度1：主动型 vs 被动型
| 类型 | 定义 | 示例 |
|---|---|---|
| 主动型 | 预判用户需求。 | 智能建议、预加载。 |
| 被动型 | 响应用户明确输入。 | 表单处理、点击操作。 |
维度2：静态型 vs 动态型
| 类型 | 更新模式 | 存储模型 |
|---|---|---|
| 静态型 | 增量式、版本化。 | 基于文件、缓存存储。 |
| 动态型 | 持续式、流式。 | 数据库、实时存储。 |
详情请见classification-system.md。
Phase 3: Scoring (feature-review:scored
)
feature-review:scored阶段3：评分（feature-review:scored
）
feature-review:scoredApply hybrid RICE+WSJF scoring:
Feature Score = Value Score / Cost Score
Value Score = (Reach + Impact + Business Value + Time Criticality) / 4
Cost Score = (Effort + Risk + Complexity) / 3
Adjusted Score = Feature Score * ConfidenceScoring Scale: Fibonacci (1, 2, 3, 5, 8, 13).
Thresholds:
- > 2.5: High priority.
- 1.5 - 2.5: Medium priority.
- < 1.5: Low priority.
See scoring-framework.md for the framework.
应用RICE+WSJF混合评分法：
功能评分 = 价值评分 / 成本评分
价值评分 = (覆盖用户数 + 影响程度 + 业务价值 + 时间紧迫性) / 4
成本评分 = (开发工作量 + 风险 + 复杂度) / 3
调整后评分 = 功能评分 * 置信度评分尺度： 斐波那契数列（1, 2, 3, 5, 8, 13）。
阈值：
- >2.5：高优先级。
- 1.5-2.5：中优先级。
- <1.5：低优先级。
详情请见scoring-framework.md。
Phase 4: Tradeoff Analysis (feature-review:tradeoffs-analyzed
)
feature-review:tradeoffs-analyzed阶段4：权衡分析（feature-review:tradeoffs-analyzed
）
feature-review:tradeoffs-analyzedEvaluate each feature across quality dimensions:
| Dimension | Question | Scale |
|---|---|---|
| Quality | Does it deliver correct results? | 1-5 |
| Latency | Does it meet timing requirements? | 1-5 |
| Token Usage | Is it context-efficient? | 1-5 |
| Resource Usage | Is CPU/memory reasonable? | 1-5 |
| Redundancy | Does it handle failures gracefully? | 1-5 |
| Readability | Can others understand it? | 1-5 |
| Scalability | Will it handle 10x load? | 1-5 |
| Integration | Does it play well with others? | 1-5 |
| API Surface | Is it backward compatible? | 1-5 |
See tradeoff-dimensions.md for criteria.
从多个质量维度评估每个功能：
| 维度 | 问题 | 评分尺度 |
|---|---|---|
| 质量 | 是否输出正确结果？ | 1-5 |
| 延迟 | 是否满足时间要求？ | 1-5 |
| Token使用 | 是否高效利用上下文？ | 1-5 |
| 资源使用 | CPU/内存占用是否合理？ | 1-5 |
| 冗余性 | 是否能优雅处理故障？ | 1-5 |
| 可读性 | 他人是否易于理解？ | 1-5 |
| 可扩展性 | 是否能支撑10倍流量？ | 1-5 |
| 集成性 | 是否能与其他功能良好协作？ | 1-5 |
| API兼容性 | 是否向后兼容？ | 1-5 |
详情请见tradeoff-dimensions.md。
Phase 5: Gap Analysis & Suggestions (feature-review:suggestions-generated
)
feature-review:suggestions-generated阶段5：差距分析与建议（feature-review:suggestions-generated
）
feature-review:suggestions-generated- Identify gaps: Missing Kano basics.
- Surface opportunities: High-value, low-effort features.
- Flag technical debt: Features with declining scores.
- Recommend actions: Build, improve, deprecate, or maintain.
- 识别差距：缺失的Kano基础功能。
- 挖掘机会：高价值、低工作量的功能。
- 标记技术债务：评分持续下降的功能。
- 推荐行动：开发、优化、弃用或维护。
Phase 6: GitHub Integration (feature-review:issues-created
)
feature-review:issues-created阶段6：GitHub集成（feature-review:issues-created
）
feature-review:issues-created- Generate issue title and body from suggestions.
- Apply labels (feature, enhancement, priority/*).
- Link to related issues.
- Confirm with user before creation.
- 从建议中生成Issue标题和内容。
- 应用标签（feature、enhancement、priority/*）。
- 关联相关Issues。
- 创建前需获得用户确认。
Configuration
配置
Feature-review uses opinionated defaults but allows customization.
Feature-review采用约定优于配置的原则，但支持自定义。
Configuration File
配置文件
Create in project root:
.feature-review.yamlyaml
undefined在项目根目录创建：
.feature-review.yamlyaml
undefined.feature-review.yaml
.feature-review.yaml
version: 1
version: 1
Scoring weights (must sum to 1.0)
评分权重（总和必须为1.0）
weights:
value:
reach: 0.25
impact: 0.30
business_value: 0.25
time_criticality: 0.20
cost:
effort: 0.40
risk: 0.30
complexity: 0.30
weights:
value:
reach: 0.25
impact: 0.30
business_value: 0.25
time_criticality: 0.20
cost:
effort: 0.40
risk: 0.30
complexity: 0.30
Score thresholds
评分阈值
thresholds:
high_priority: 2.5
medium_priority: 1.5
thresholds:
high_priority: 2.5
medium_priority: 1.5
Tradeoff dimension weights (0.0 to disable)
权衡维度权重（设为0.0则禁用）
tradeoffs:
quality: 1.0
latency: 1.0
token_usage: 1.0
resource_usage: 0.8
redundancy: 0.5
readability: 1.0
scalability: 0.8
integration: 1.0
api_surface: 1.0
See [configuration.md](modules/configuration.md) for options.tradeoffs:
quality: 1.0
latency: 1.0
token_usage: 1.0
resource_usage: 0.8
redundancy: 0.5
readability: 1.0
scalability: 0.8
integration: 1.0
api_surface: 1.0
详情请见[configuration.md](modules/configuration.md)。Guardrails
防护规则
These rules apply to all configurations:
- Minimum dimensions: Evaluate at least 5 tradeoff dimensions.
- Confidence requirement: Review scores below 50% confidence.
- Breaking change warning: Require acknowledgment for API surface changes.
- Backlog limit: Limit suggestion queue to 25 items.
所有配置需遵循以下规则：
- 最小维度要求：至少评估5个权衡维度。
- 置信度要求：需重新评审置信度低于50%的评分。
- 破坏性变更警告：API变更需获得确认。
- 待办事项限制：建议队列最多25项。
Required TodoWrite Items
必填TodoWrite项
feature-review:inventory-completefeature-review:classifiedfeature-review:scoredfeature-review:tradeoffs-analyzedfeature-review:suggestions-generated- (if requested)
feature-review:issues-created
feature-review:inventory-completefeature-review:classifiedfeature-review:scoredfeature-review:tradeoffs-analyzedfeature-review:suggestions-generated- （若启用）
feature-review:issues-created
Integration Points
集成点
- : Provides Worthiness Scores for suggestions.
imbue:scope-guard - : Prioritizes issues with high scores.
sanctum:do-issue - : Evaluates new ideas against existing features.
superpowers:brainstorming
- ：为建议提供价值评分。
imbue:scope-guard - ：为高分建议优先创建Issues。
sanctum:do-issue - ：对照现有功能评估新想法。
superpowers:brainstorming
Output Format
输出格式
Feature Inventory Table
功能盘点表格
markdown
| Feature | Type | Data | Score | Priority | Status |
|---------|------|------|-------|----------|--------|
| Auth middleware | Reactive | Dynamic | 2.8 | High | Stable |
| Skill loader | Reactive | Static | 2.3 | Medium | Needs improvement |markdown
| 功能 | 类型 | 数据属性 | 评分 | 优先级 | 状态 |
|---------|------|------|-------|----------|--------|
| 认证中间件 | 被动型 | 动态 | 2.8 | 高 | 稳定 |
| 技能加载器 | 被动型 | 静态 | 2.3 | 中 | 需要优化 |Suggestion Report
建议报告
markdown
undefinedmarkdown
undefinedFeature Suggestions
功能建议
High Priority (Score > 2.5)
高优先级（评分>2.5）
- [Feature Name] (Score: 2.7)
- Classification: Proactive/Dynamic
- Value: High reach
- Cost: Moderate effort
- Recommendation: Build in next sprint
undefined- [功能名称]（评分：2.7）
- 分类：主动型/动态
- 价值：覆盖用户数多
- 成本：工作量中等
- 建议：下一个迭代开发
undefinedRelated Skills
相关技能
- : Prevent overengineering.
imbue:scope-guard - : Structured review methodology.
imbue:review-core - : Code-level feature review.
sanctum:pr-review
- ：防止过度设计。
imbue:scope-guard - ：结构化评审方法论。
imbue:review-core - ：代码级功能评审。
sanctum:pr-review
Reference
参考资料
- scoring-framework.md: RICE+WSJF hybrid.
- classification-system.md: Axes definition.
- tradeoff-dimensions.md: Quality attributes.
- configuration.md: Customization options.
- scoring-framework.md：RICE+WSJF混合评分法。
- classification-system.md：分类维度定义。
- tradeoff-dimensions.md：质量属性说明。
- configuration.md：自定义选项。
Troubleshooting
故障排查
Common Issues
常见问题
Command not found
Ensure all dependencies are installed and in PATH
Permission errors
Check file permissions and run with appropriate privileges
Unexpected behavior
Enable verbose logging with flag
--verbose命令未找到
确保所有依赖已安装并添加至PATH
权限错误
检查文件权限并使用合适权限运行
异常行为
使用 flag启用详细日志
--verbose