full-stack-master

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Full Stack Master Workflow Skill

一、能力定位 (Capability)

1. Capability Positioning

工作流自动编排：串联需求设计开发测试质量文档提交审核发布的全链路。
技能聚合：集成所有核心原子技能（Todo Manager、Technical Planning、Backend Logic Expert、Vue Frontend Expert、Database Expert、Code Quality Auditor、Test Engineer、UI Validator、Conventional Committer）。
可复用与可拓展：可合并新场景（如数据库迁移、API 变更、运营发布等），支持多项目切换。
分阶段接棒/派单：可手动或脚本分配阶段任务给对应技能或专项 agent。

Workflow Automatic Orchestration: Connect the entire lifecycle of requirement design, development, testing, quality assurance, documentation, submission, review, and release.
Skill Aggregation: Integrate all core atomic skills (Todo Manager, Technical Planning, Backend Logic Expert, Vue Frontend Expert, Database Expert, Code Quality Auditor, Test Engineer, UI Validator, Conventional Committer).
Reusable and Expandable: Can incorporate new scenarios (such as database migration, API changes, operation release, etc.) and support multi-project switching.
Phased Handover/Task Assignment: Manually or script-based assignment of phase tasks to corresponding skills or specialized agents.

二、强制参考文档 (Mandatory Documentation)

2. Mandatory Documentation

在执行任何写操作或决策前，必须确保已读取并理解以下文档的最新内容：

全周期基石：AGENTS.md (安全红线与身份)、AI 协作规范
规划与任务：项目规划、待办事项、项目规划规范
开发与设计：开发规范、API 规范、UI 设计
安全与质量：安全规范、测试规范

Before performing any write operation or making decisions, you must ensure that you have read and understood the latest content of the following documents:

Lifecycle Foundation: AGENTS.md (Security Red Lines and Identity), AI Collaboration Standards
Planning and Tasks: Project Roadmap, To-Do Items, Project Planning Standards
Development and Design: Development Standards, API Standards, UI Design
Security and Quality: Security Standards, Testing Standards

三、标准 PDTFC+ 2.0 工作流 (Standard Workflow)

3. Standard PDTFC+ 2.0 Workflow

P (Plan) - 需求分析与规划
1. 读取文档：确认
```
todo.md
```
  和
```
roadmap.md
```
  。
2. 意图抽离：启动采访追问程序同步需求。
3. 方案设计：使用
```
technical-planning
```
  规划改动清单，并使用
```
todo-manager
```
  更新状态。
- 技能：
```
requirement-analyst
```
  、
```
todo-manager
```
  、
```
technical-planning
```
D (Do) - 开发实现
1. 核心实现：遵循开发规范；若涉及持久化，优先开发
```
database-expert
```
  实体。
2. 自检修复：开发完成通过
```
code-quality-auditor
```
  消除 Lint 和类型报错。
- 技能：
```
database-expert
```
  、
```
backend-logic-expert
```
  、
```
vue-frontend-expert
```
  、
```
code-quality-auditor
```
A (Audit) - 代码审计
1. 安全审计：扫描注入、越权与敏感信息。
2. 规范审计：对比
```
todo.md
```
  确认功能点与规划一致。
- 技能：
```
code-quality-auditor
```
  、
```
security-guardian
```
C1 (Commit) - 功能提交
1. 原子提交：使用
```
conventional-committer
```
  执行第一次提交。
V (Validate) - UI 验证
1. 视觉准则：浏览器验证实际渲染效果。若自动化工具失效，应向用户展示截图或请求人工验证。
- 技能：
```
ui-validator
```
T (Test) - 自动化测试
1. 定向测试：编写并运行 Vitest 用例（测试代码也需过审计）。
- 技能：
```
test-engineer
```
C2 (Commit) - 测试提交
1. 最终交付：提交测试代码。
F (Finish) - 任务完结
1. 文档更新：更新
```
todo.md
```
  状态，并归并项目变更文档。
- 技能：
```
todo-manager
```
  、
```
documentation-specialist
```
质量检测与审查 (Test/Review)
- 要求：执行测试前，必须读取 测试规范。
- 任务：运行
```
pnpm lint
```
  ,
```
pnpm typecheck
```
  以及定向/按需测试。
- 策略: 遵循高效测试策略，除非必要否则不执行全量测试。
- 技能：
```
quality-guardian
```
  、
```
test-engineer
```
  、
```
code-reviewer
```
问题修复 (Fix)
- 目标：消除上阶段发现的所有缺陷。
- 技能：
```
nuxt-code-editor
```
功能提交 (Commit - Phase 1)
- 目标：在通过核心质量检查后提交业务逻辑。
- 任务：使用 Conventional Commits 规范（中文）提交。
- 要求: 遵循提交规模与原子化改动，确保一次提交对应一个 Todo，文件数 < 10。
- 技能：
```
conventional-committer
```
测试增强 (Enhance)
- 目标：补齐测试用例，提升代码覆盖率。
- 任务：为新功能补齐正向、反向及边缘场景的 Vitest 用例。
- 技能：
```
test-engineer
```
测试提交 (Commit - Phase 2)
- 目标：将增强后的测试代码入库。
- 技能：
```
conventional-committer
```

P (Plan) - Requirement Analysis and Planning
1. Read Documents: Confirm the content of
```
todo.md
```
  and
```
roadmap.md
```
  .
2. Intent Extraction: Launch the interview and follow-up process to synchronize requirements.
3. Solution Design: Use
```
technical-planning
```
  to plan the change list, and update the status with
```
todo-manager
```
  .
- Skills:
```
requirement-analyst
```
  ,
```
todo-manager
```
  ,
```
technical-planning
```
D (Do) - Development Implementation
1. Core Implementation: Follow the Development Standards; if persistence is involved, prioritize developing entities with
```
database-expert
```
  .
2. Self-Inspection and Fixing: After development, eliminate Lint and type errors through
```
code-quality-auditor
```
  .
- Skills:
```
database-expert
```
  ,
```
backend-logic-expert
```
  ,
```
vue-frontend-expert
```
  ,
```
code-quality-auditor
```
A (Audit) - Code Audit
1. Security Audit: Scan for injection, privilege escalation, and sensitive information leaks.
2. Specification Audit: Compare with
```
todo.md
```
  to confirm that function points are consistent with the plan.
- Skills:
```
code-quality-auditor
```
  ,
```
security-guardian
```
C1 (Commit) - Feature Submission
1. Atomic Commit: Use
```
conventional-committer
```
  to perform the first commit.
V (Validate) - UI Validation
1. Visual Guidelines: Verify the actual rendering effect in the browser. If automated tools fail, display screenshots to the user or request manual verification.
- Skills:
```
ui-validator
```
T (Test) - Automated Testing
1. Targeted Testing: Write and run Vitest cases (test code must also pass the audit).
- Skills:
```
test-engineer
```
C2 (Commit) - Test Submission
1. Final Delivery: Submit the test code.
F (Finish) - Task Completion
1. Document Update: Update the status in
```
todo.md
```
  and merge project change documents.
- Skills:
```
todo-manager
```
  ,
```
documentation-specialist
```
Quality Inspection and Review (Test/Review)
- Requirements: Before conducting tests, you must read the Testing Standards.
- Tasks: Run
```
pnpm lint
```
  ,
```
pnpm typecheck
```
  and targeted/on-demand testing.
- Strategy: Follow the Efficient Testing Strategy, and do not perform full-scale testing unless necessary.
- Skills:
```
quality-guardian
```
  ,
```
test-engineer
```
  ,
```
code-reviewer
```
Issue Fixing (Fix)
- Goal: Eliminate all defects found in the previous stage.
- Skills:
```
nuxt-code-editor
```
Feature Submission (Commit - Phase 1)
- Goal: Submit the business logic after passing the core quality inspection.
- Task: Submit following the Conventional Commits specification (Chinese version).
- Requirements: Follow Commit Scale & Atomic Changes, ensuring one commit corresponds to one Todo item and the number of files is less than 10.
- Skills:
```
conventional-committer
```
Test Enhancement (Enhance)
- Goal: Complete test cases to improve code coverage.
- Task: Add positive, negative, and edge scenario Vitest cases for new features.
- Skills:
```
test-engineer
```
Test Submission (Commit - Phase 2)
- Goal: Store the enhanced test code in the repository.
- Skills:
```
conventional-committer
```

四、需求挖掘方法论 (Intent Extraction Methodology)

4. Intent Extraction Methodology

逐级递进：先锁定整体结构和目标，再深入到具体实现细节。
单点突破：一次仅问一个问题，待用户回答后再进行下一步追问。
循环校验：当用户回答不清晰时，尝试换一种表述方式进行确认。
意图抽离：分析用户想要什么背后的为什么，提供更优专业建议。

Gradual Progression: First lock the overall structure and goals, then delve into specific implementation details.
Single Point Breakthrough: Ask only one question at a time, and proceed to the next follow-up after receiving the user's answer.
Cycle Verification: If the user's answer is unclear, try rephrasing to confirm.
Intent Extraction: Analyze the "why" behind what the user wants, and provide better professional suggestions.

五、技能引用（Each Sub-Skill Reference）

5. Each Sub-Skill Reference

context-analyzer
nuxt-code-editor
test-engineer
quality-guardian
documentation-specialist
code-reviewer
conventional-committer
ui-validator

context-analyzer
nuxt-code-editor
test-engineer
quality-guardian
documentation-specialist
code-reviewer
conventional-committer
ui-validator

六、编写规范 (Authoring Rules)

6. Authoring Rules

Imperative & Structured
- 用动词+目标描述标准化每一步/每个技能的 usage section。
- 禁止冗长废话和流程介绍型文字。
明确输入输出
- 每步须说明本阶段输入依赖、输出产物（如文件路径、文档链接）。
- 例：输入：docs/plan/，输出：docs/design/xx.md。
可链式组合
- 每步技能应允许独立、或作为全局 master 调用链局部片段。
- 部分技能支持多角色协同（如测试、文档可并行）。
安全检查与通用异常处理
- 强行插入 typecheck、lint 等质量关卡，禁止在未检测前进入提交/发布环节。
- 明确安全等级和数据保护点。
国际化与文档优先
- 所有工作流/技能创建应默认兼容 i18n 和标准文档同步动作。

Imperative & Structured
- Standardize the usage section of each step/skill with verb + target description.
- Prohibit lengthy nonsense and process introduction text.
Clear Input and Output
- Each step must specify the input dependencies and output products (such as file paths, document links).
- Example: Input: docs/plan/, Output: docs/design/xx.md.
Chainable Combination
- Each step skill should allow independent use or serve as a local fragment of the global master call chain.
- Some skills support multi-role collaboration (e.g., testing and documentation can be done in parallel).
Security Check and General Exception Handling
- Mandatorily insert quality checkpoints such as typecheck and lint, and prohibit entering the submission/release stage without detection.
- Clarify security levels and data protection points.
Internationalization and Documentation-First
- All workflows/skills should be created with default compatibility for i18n and standard document synchronization actions.

七、模板用法 (Usage Example)

7. Usage Example

yaml

workflow:
  - step: "需求分析"        # context-analyzer, documentation-specialist
  - step: "功能开发"        # nuxt-code-editor
  - step: "UI 验证"         # ui-validator
  - step: "质量检测"        # quality-guardian, code-reviewer
  - step: "功能提交"        # conventional-committer
  - step: "测试补充"        # test-engineer
  - step: "测试提交"        # conventional-committer

yaml

workflow:
  - step: "需求分析"        # context-analyzer, documentation-specialist
  - step: "功能开发"        # nuxt-code-editor
  - step: "UI 验证"         # ui-validator
  - step: "质量检测"        # quality-guardian, code-reviewer
  - step: "功能提交"        # conventional-committer
  - step: "测试补充"        # test-engineer
  - step: "测试提交"        # conventional-committer