用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/AsiaOstrich/universal-dev-standards --skill discover命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | discover |
| scope | universal |
| description | [UDS] Assess project health, architecture, and risks before adding features |
| allowed-tools | Read, Grep, Glob, Bash(npm test:*), Bash(npm audit:*), Bash(npm outdated:*) |
| argument-hint | [feature area | 功能範圍] |
| disable-model-invocation | true |
Phase 0 assessment before adding features to existing codebases. Evaluate project health, architecture, and risks.
在既有程式碼庫新增功能前的 Phase 0 評估。評估專案健康度、架構與風險。
| Dimension | What to Check | 檢查項目 |
|---|---|---|
| Architecture | Module structure, dependency graph, entry points | 模組結構、相依圖、進入點 |
| Dependencies | Outdated packages, known vulnerabilities, license risks | 過時套件、已知漏洞、授權風險 |
| Test Coverage | Existing test suite, coverage gaps, test quality | 現有測試、覆蓋率缺口、測試品質 |
| Security | npm audit findings, hardcoded secrets, exposed endpoints | 安全稽核、硬編碼密鑰、暴露端點 |
| Technical Debt | TODOs, code duplication, complexity hotspots | TODO 標記、程式碼重複、複雜度熱點 |
npm outdated, npm audit for health signalsProject Health Report
=====================
Overall Score: 7.2 / 10
| Dimension | Score | Status | Key Finding |
|-----------------|-------|---------|------------------------|
| Architecture | 8/10 | Good | Clean module boundaries |
| Dependencies | 6/10 | Warning | 5 outdated, 1 critical |
| Test Coverage | 7/10 | Fair | 72% line coverage |
| Security | 8/10 | Good | No critical vulns |
| Technical Debt | 6/10 | Warning | 23 TODOs, 3 hotspots |
Recommendations:
1. [HIGH] Update lodash to fix CVE-2024-XXXX
2. [MED] Add tests for src/payments/ (0% coverage)
3. [LOW] Resolve TODO backlog in src/utils/
After assessment, identified risks are recorded in a structured Risk Register for ongoing tracking.
評估完成後，已識別的風險記錄在結構化的風險登記簿中進行持續追蹤。
Impact | 影響
Low Med High
High [ Med ] [High] [Crit]
Med [ Low ] [Med ] [High] Likelihood | 可能性
Low [ Low ] [Low ] [Med ]
# Risk Register — [Project Name]
**Last Updated**: YYYY-MM-DD
| ID | Category | Description | Likelihood | Impact | Level | Owner | Mitigation | Status |
|----|----------|-------------|-----------|--------|-------|-------|------------|--------|
| RISK-001 | Security | Outdated deps with CVEs | High | High | Critical | @dev | npm audit fix | Open |
| RISK-002 | Performance | No load testing | Medium | High | High | @ops | Add k6 tests | Open |
| RISK-003 | Quality | Low test coverage in payments | High | Medium | High | @qa | Add IT tests | Mitigating |
Identified ──► Mitigating ──► Resolved ──► Closed
│
└──► Accepted (with justification)
docs/risks/
├── RISK-REGISTER.md # Active risk register
├── RISK-REGISTER-2026-Q1.md # Quarterly snapshot (optional)
└── README.md # Index
/discover - Full project health assessment/discover auth - Focused assessment of auth-related modules/discover payments - Assess risks before adding payment features/discover --risks - View current risk register/discover --update-risk RISK-NNN - Update a risk item statusAfter /discover completes, the AI assistant should suggest based on the assessment:
根據評估結果，建議下一步 / Based on assessment, suggested next steps:
- New feature / 新功能 →
/sddto create a specification ⭐ Recommended / 推薦- Legacy code / 遺留程式碼 →
/reverse specto extract existing behavior- Refactoring / 重構 →
/refactor decideto choose a strategy- Quick fix / 快速修復 →
/tddto write a targeted test and fix- Risk tracking / 風險追蹤 →
/discover --risksto view risk register- Architecture decision / 架構決策 →
/adrto record decisions made during discovery
完整的 AI 行為定義請參閱對應的命令文件：
/discoverFor complete AI agent behavior definition, see the corresponding command file:
/discover
[UDS] 以 Claude 原生 Agent tool 编排多任务执行计划（DAG-based，无外部引擎）。 Use when: executing a plan.json file with parallel/sequential task dependencies. Keywords: orchestrate, plan, execute, DAG, task plan, 编排, 执行计划, 并行.
[UDS] 从 Spec 文档、OpenSpec 变更或自由文本需求生成 plan.json。 Use when: converting specifications into executable task plans for /orchestrate. Keywords: plan, spec, task plan, 计划, 规格, 任务, plan.json, DAG.
[UDS] AI 辅助 git push 安全层：质量门禁 + 协作护栏。 Use when: pushing commits, force pushing, pushing to protected branches, pushing feature branches. Keywords: git push, force push, protected branch, quality gate, push receipt, PR automation, 推送, 保护分支, 质量门禁.