Back to Details

executing-plans

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

执行计划

Execution Plan

概述

Overview

加载计划，批判性审查，执行所有任务，完成后报告。
开始时宣布： "我正在使用 executing-plans 技能来实现此计划。"
注意： 告诉你的人类伙伴，Superpowers 在有子代理支持时效果好得多。如果在支持子代理的平台上运行（如 Claude Code 或 Codex），其工作质量会显著提高。如果子代理可用，请使用 superpowers:subagent-driven-development 而非此技能。
Load the plan, conduct a critical review, execute all tasks, and report upon completion.
Announce at the start: "I am using the executing-plans skill to implement this plan."
Note: Tell your human partner that Superpowers work much better with subagent support. When running on platforms that support subagents (such as Claude Code or Codex), the quality of work improves significantly. If subagents are available, use superpowers:subagent-driven-development instead of this skill.

流程

Process

步骤 1：加载并审查计划

Step 1: Load and Review the Plan

  1. 读取计划文件
  2. 批判性审查——识别计划中的任何问题或疑虑
  3. 如果有疑虑：在开始之前向你的人类伙伴提出
  4. 如果没有疑虑：创建 TodoWrite 并继续
审查时重点检查：
  • 步骤之间是否有依赖遗漏？（A 依赖 B，但 B 排在 A 之后）
  • 验证条件是否明确？（"确认可用"不算，"运行 
    npm test
    全部通过"才算）
  • 是否有隐含的环境假设？（Node 版本、数据库连接、API Key）
审查示例：
计划文件：docs/plan.md
任务清单：5 个任务

审查发现：
- 任务 3（添加数据库迁移）应在任务 2（编写数据模型）之后，顺序正确 ✓
- 任务 4 的验证条件写的是"确认功能正常"→ 需澄清：具体跑什么测试？
- 计划未提及 Python 版本要求 → 需确认

向伙伴提出：
"计划整体可执行。有两个问题：(1) 任务 4 的验证条件不够具体，建议改为
'运行 pytest tests/test_api.py 全部通过'；(2) 需要确认 Python 版本要求。"
  1. Read the plan file
  2. Critical review — identify any issues or concerns in the plan
  3. If there are concerns: Raise them with your human partner before starting
  4. If no concerns: Create a TodoWrite and proceed
Key review points:
  • Are there missing dependencies between steps? (A depends on B, but B is scheduled after A)
  • Are validation conditions clear? ("Confirm availability" is not acceptable; "Run 
    npm test
    with all passing" is)
  • Are there implicit environment assumptions? (Node version, database connection, API Key)
Review example:
Plan file: docs/plan.md
Task list: 5 tasks

Review findings:
- Task 3 (Add database migration) should come after Task 2 (Write data model), order is correct ✓
- Validation condition for Task 4 is written as "Confirm function works" → Needs clarification: What specific tests should be run?
- Plan does not mention Python version requirements → Needs confirmation

Raise to partner:
"The plan is generally executable. There are two issues: (1) The validation condition for Task 4 is not specific enough, it is recommended to change it to
'Run pytest tests/test_api.py with all passing'; (2) Python version requirements need to be confirmed."

步骤 2：执行任务

Step 2: Execute Tasks

对于每个任务：
  1. 标记为进行中 — 更新 TodoWrite
  2. 理解目标 — 重读任务描述，明确完成标准
  3. 执行实现 — 严格按照计划步骤执行（计划已有小步骤）
  4. 运行验证 — 按要求运行测试或检查
  5. 提交变更 — 每完成一个任务提交一次，commit message 引用任务编号
  6. 标记为已完成 — 更新 TodoWrite
每个任务的节奏：
--- 任务 2/5：添加用户验证 ---
[标记进行中]

目标：为 /api/users 添加输入验证
完成标准：所有验证测试通过，无效输入返回 400

[实现]
- 添加 validateUser() 中间件
- 编写 3 个验证规则（email 格式、密码强度、用户名长度）

[验证]
$ npm test -- --grep "validation"
  ✓ 拒绝无效 email (12ms)
  ✓ 拒绝弱密码 (8ms)
  ✓ 拒绝过长用户名 (5ms)
  3 passing

[提交]
$ git add src/middleware/validate.js tests/validation.test.js
$ git commit -m "feat: 添加用户输入验证（任务 2/5）"

[标记完成]
--- 任务 2/5 完成 ---
批量审查检查点：
  • 每完成 3 个任务后，暂停回顾：整体方向还对吗？有没有偏离计划？
  • 如果发现前面的实现有问题，先修复再继续，不要带着问题往下走
For each task:
  1. Mark as in progress — Update TodoWrite
  2. Understand the goal — Re-read the task description to clarify completion criteria
  3. Execute implementation — Strictly follow the plan steps (the plan already includes small steps)
  4. Run validation — Execute required tests or checks as instructed
  5. Submit changes — Commit once per completed task, reference the task number in the commit message
  6. Mark as completed — Update TodoWrite
Rhythm for each task:
--- Task 2/5: Add user validation ---
[Mark as in progress]

Goal: Add input validation for /api/users
Completion criteria: All validation tests pass, invalid inputs return 400

[Implementation]
- Add validateUser() middleware
- Write 3 validation rules (email format, password strength, username length)

[Validation]
$ npm test -- --grep "validation"
  ✓ Reject invalid email (12ms)
  ✓ Reject weak password (8ms)
  ✓ Reject too long username (5ms)
  3 passing

[Commit]
$ git add src/middleware/validate.js tests/validation.test.js
$ git commit -m "feat: Add user input validation (Task 2/5)"

[Mark as completed]
--- Task 2/5 Completed ---
Batch review checkpoints:
  • After completing every 3 tasks, pause and review: Is the overall direction still correct? Have we deviated from the plan?
  • If issues are found in previous implementations, fix them before proceeding; do not continue with unresolved problems

步骤 3：处理常见异常

Step 3: Handle Common Exceptions

测试失败：
  1. 读错误信息，定位失败原因
  2. 区分：是实现 bug？还是测试本身有问题？还是计划描述有误？
  3. 实现 bug → 修复并重跑
  4. 测试有问题 → 修复测试，向伙伴说明
  5. 计划有误 → 停下来，向伙伴报告并建议修正
依赖缺失：
任务 3 需要 Redis 连接，但计划中没有提及 Redis 配置。
→ 停止执行
→ 向伙伴报告："任务 3 需要 Redis，计划中未包含配置步骤。
   建议：在任务 3 前插入 '配置 Redis 连接' 步骤。"
指令不清：
  • 不要猜测意图，不要"合理推断"
  • 列出你的理解和困惑，让伙伴澄清
  • 等待回复后再继续
Test failure:
  1. Read the error message to locate the cause of failure
  2. Differentiate: Is it an implementation bug? A problem with the test itself? Or an error in the plan description?
  3. Implementation bug → Fix and re-run
  4. Test problem → Fix the test and explain to the partner
  5. Plan error → Stop execution, report to the partner and suggest corrections
Missing dependencies:
Task 3 requires a Redis connection, but the plan does not mention Redis configuration.
→ Stop execution
→ Report to partner: "Task 3 requires Redis, but the plan does not include configuration steps.
   Suggestion: Insert 'Configure Redis connection' step before Task 3."
Unclear instructions:
  • Do not guess intent, do not "reasonably infer"
  • List your understanding and confusion, ask the partner to clarify
  • Wait for a response before proceeding

步骤 4：完成开发

Step 4: Complete Development

所有任务完成并验证后：
  • 宣布："我正在使用 finishing-a-development-branch 技能来完成此工作。"
  • 必需子技能： 使用 superpowers:finishing-a-development-branch
  • 按照该技能的指引验证测试、展示选项、执行选择
完成报告模板：
undefined
After all tasks are completed and validated:
  • Announce: "I am using the finishing-a-development-branch skill to complete this work."
  • Required sub-skill: Use superpowers:finishing-a-development-branch
  • Follow the guidance of that skill to validate tests, present options, and execute selections
Completion report template:
undefined

执行报告

Execution Report

计划： docs/plan.md 分支： feature/user-validation 任务： 5/5 已完成
Plan: docs/plan.md Branch: feature/user-validation Tasks: 5/5 Completed

完成的任务

Completed Tasks

  1. ✅ 初始化项目结构
  2. ✅ 添加用户验证
  3. ✅ 添加数据库迁移
  4. ✅ 实现 API 端点
  5. ✅ 添加集成测试
  1. ✅ Initialize project structure
  2. ✅ Add user validation
  3. ✅ Add database migration
  4. ✅ Implement API endpoints
  5. ✅ Add integration tests

验证结果

Validation Results

  • 单元测试：23/23 通过
  • 集成测试：8/8 通过
  • lint 检查：0 个警告
  • Unit tests: 23/23 Passed
  • Integration tests: 8/8 Passed
  • Lint check: 0 warnings

偏离计划的地方

Deviations from Plan

  • 任务 3：Redis 配置从 env 改为 config.yaml（经伙伴同意）
  • Task 3: Redis configuration changed from env to config.yaml (with partner's approval)

下一步

Next Steps

按 finishing-a-development-branch 技能处理合并/PR
undefined
Process merge/PR according to the finishing-a-development-branch skill
undefined

何时停下来求助

When to Stop and Ask for Help

在以下情况立即停止执行：
  • 遇到阻塞（缺少依赖、测试失败、指令不清）
  • 计划有严重缺陷导致无法开始
  • 你不理解某条指令
  • 验证反复失败（同一测试失败 2 次以上）
不确定时就问，不要猜测。
Stop execution immediately in the following cases:
  • Encountering blocks (missing dependencies, test failures, unclear instructions)
  • The plan has serious flaws that prevent starting
  • You do not understand a certain instruction
  • Validation fails repeatedly (same test fails more than 2 times)
Ask when in doubt, do not guess.

何时回到之前的步骤

When to Return to Previous Steps

回到审查（步骤 1）当：
  • 伙伴根据你的反馈更新了计划
  • 根本性的方案需要重新考虑
不要硬闯阻塞 — 停下来问。
Return to review (Step 1) when:
  • The partner updates the plan based on your feedback
  • The fundamental solution needs to be reconsidered
Do not force through blocks — Stop and ask.

注意事项

Notes

  • 先批判性审查计划
  • 严格按照计划步骤执行
  • 不要跳过验证
  • 每个任务单独提交，commit message 引用任务编号
  • 计划要求时引用相应技能
  • 遇到阻塞时停下来，不要猜测
  • 未经用户明确同意，绝不在 main/master 分支上开始实现
  • Conduct a critical review of the plan first
  • Strictly follow the plan steps
  • Do not skip validation
  • Commit separately for each task, reference the task number in the commit message
  • Reference corresponding skills when required by the plan
  • Stop when encountering blocks, do not guess
  • Never start implementation on the main/master branch without explicit user approval

集成

Integration

必需的工作流技能：
  • superpowers:using-git-worktrees - 必需：开始前建立隔离的工作空间
  • superpowers:writing-plans - 创建此技能要执行的计划
  • superpowers:finishing-a-development-branch - 所有任务完成后收尾开发
Required workflow skills:
  • superpowers:using-git-worktrees - Required: Set up an isolated workspace before starting
  • superpowers:writing-plans - Create plans to be executed by this skill
  • superpowers:finishing-a-development-branch - Wrap up development after all tasks are completed