analysis-delivery-workflow — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited analysis-delivery-workflow (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.
paths/*.mdHANDOVER.mdtask-confirm-check.py --strict 用于 2→3;QA P0=0 才能 8→9stage-gate, doc-numbering, context-pointer, goal-boundaryknowledge-path, tech-stack-path, compliance-path, doc-naming-path/setup-analysis-delivery,再按顺序走 9 阶段;阶段 9 之后可选 /using-superpowers| # | 阶段 | 调用的 skill | 产出 |
|---|---|---|---|
| 1 | 项目配置 | /setup-analysis-delivery | 4 个 *-path.md |
| 2 | 需求澄清 | /grill-task | TASK_CONFIRM + 字段对齐 |
| 3 | BRD | /to-brd | 01-BRD + 流程图 |
| 4 | 合规评审 | /compliance-review | 04-合规评审 |
| 5 | 测试用例 | /test-case-design | 07-测试用例 |
| 6 | PRD | /to-prd | 05-PRD(三格式) |
| 7 | 开发设计 | /dev-design | AGENTS + FSD + 数据模型 + 开发设计 + 回测 + 复盘 |
| 8 | QA 审计 | /qa-audit | 09-QA 审计报告 |
| 9 | 交接 | /handoff | HANDOVER.md |
设计交接完成后,推荐进入 `/using-superpowers` 走 5 步实施:
brainstorming → design-an-interface → domain-modeling → writing-plans → tdd → executing-plans → verification-before-completion| 实施步骤 | 产出 |
|---|---|
| brainstorming | 设计稿 |
| design-an-interface | 接口契约 |
| domain-modeling | 领域模型 |
| writing-plans | ≤ 2h 子任务列表 |
| tdd | RED → GREEN → REFACTOR |
| executing-plans | 逐步开发 + commit |
| verification-before-completion | 完成前验证 |
每阶段必签字才能进入下一阶段。详细见 disciplines/stage-gate 3 层门控:
阶段 2 → 阶段 3 之间，必须同时满足：
skills/disciplines/stage-gate/SKILL.md）详见 scripts/task-confirm-check.py 自动化校验。
每个项目根维护 .analysis-delivery-state.json(加入 .gitignore),5 个子命令:
# 初始化(项目级)
python3 scripts/analysis-state.py init --project <name> [--project-root .] [--force]
# 记录 gate 脚本结果(pass/fail 会自动 +1 拦截次数)
python3 scripts/analysis-state.py record-gate --stage N --script <gate> --result pass|fail
# 记录阶段签字(只接受 4 句白名单之一,其他一律 reject)
python3 scripts/analysis-state.py signoff --stage N --text "我已全部确认,可以进入下一步"
# 查看当前状态(9 阶段进度 + last_gate + errors)
python3 scripts/analysis-state.py status
# 5 项度量指标(total_gates / total_signoffs / 拦截次数 / 重试次数 / 阶段用时)
python3 scripts/analysis-state.py metrics [--json]白名单话术(只 4 句):
❌ "OK / 好 / 继续 / 确认" 一律视为非签字。
中断恢复:状态文件在 .gitignore,存在本地;下次开会话时跑 status 即可看到当前阶段、已签字阶段、错误计数。
rules/stage-gate — 3 层门控rules/no-field-guessing — 字段名rules/doc-numbering — 文档编号~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.