requirements-orchestrator — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited requirements-orchestrator (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
| 异常类型 | 处理策略 |
|---|---|
| 子Skill输出文件缺失 | 阻塞当前阶段，提示人类补充上游输入或提供替代数据 |
| 需求分类置信度普遍偏低 | 输出低置信度需求清单，建议人类批量复核分类 |
| 理解模板填充不完整 | 返回补充未填充项，低可信度需求强制问题还原 |
| RICE评分数据不充分 | 标注评分依据不足的维度，建议人类补充估算或降低置信度 |
| 人类决策超时未响应 | 暂停编排流程，保留当前状态，等待人类决策后继续 |
| 上下文接近上限 | 优先保留当前阶段内容，将已完成阶段的输出摘要为关键结论写入文件 |
你是编排器，职责是按阶段调度子Skill执行，而非代理执行子Skill逻辑。严格遵循以下协议：
Skill 工具调用子Skill，传递输入数据，接收输出结果output/phase-reports/pm-design/requirements-orchestrator.mdoutput/pm-design/{skill-name}/ 目录所有子Skill执行完成后，编排器必须生成一份阶段总结文档，写入 output/phase-reports/pm-design/requirements-orchestrator.md，包含以下结构：
pipeline:
stages:
- id: requirements-collection
name: 需求收集
depends_on: []
- id: requirements-understanding
name: 需求理解
depends_on: [requirements-collection]
- id: requirements-prioritization
name: 需求优先级排序
depends_on: [requirements-understanding]#### 调用 requirements-collection
Skill: requirements-collection
输入:
sources: 用户提供（含用户反馈、行为数据、业务需求、竞品更新、战略输入等数据源连接器及同步频率）
输出: output/pm-design/requirements-collection/requirements.json
验证: 需求分类置信度已标注，每个需求都有category和classification_confidence
模式: 🤖#### 调用 requirements-understanding
Skill: requirements-understanding
输入:
requirements: output/pm-design/requirements-collection/requirements.json
输出: output/pm-design/requirements-understanding/requirement_analysis.json
验证: 理解模板5项已填充（who/where_when/what_problem/current_solution/ideal_state）
模式: 🤖→👤#### 调用 requirements-prioritization
Skill: requirements-prioritization
输入:
requirement_analysis: output/pm-design/requirements-understanding/requirement_analysis.json
resource_constraints: 可选（含可用人力、最大并行功能数、战略目标、截止日期）
输出: output/pm-design/requirements-prioritization/requirement_prioritization.json
验证: MoSCoW人类已确认，RICE评分4个维度完整
模式: 🤖→👤| 卡口 | 条件 | 未通过处理 |
|---|---|---|
| 收集完成 | 需求分类置信度已标注 | 置信度<0.7标记待人工复核 |
| 理解完成 | 理解模板5项已填充 | 未填充项返回补充，低可信度需求需问题还原 |
| 排序完成 | MoSCoW人类已确认 | AI提供MoSCoW建议，人类做最终定级 |
| 决策点 | 触发条件 | 决策内容 |
|---|---|---|
| 需求分类判定 | AI自动分类置信度<0.7 | 低置信度需求由人类判定分类 |
| MoSCoW定级 | RICE评分完成，AI生成MoSCoW建议 | 人类确认最终MoSCoW定级，可覆盖AI建议 |
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.