Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问clear审计异常

boy-scout-rule童子军规则

Agent Skill

boy-scout-rule 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

552

周安装

23

GitHub Stars

142

下载量

184
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:boy-scout-rule(童子军规则)
来源仓库:https://github.com/thebushidocollective/han
仓库路径:skills/boy-scout-rule
安装命令:
npx skills add https://github.com/thebushidocollective/han --skill boy-scout-rule
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/thebushidocollective/han --skill boy-scout-rule

简介

Boy Scout Rule 鼓励每次修改代码时都使其比原来更好。

  • 适用于希望逐步改善代码质量的日常开发场景。
  • 可移除死代码、修复命名问题、添加类型注解或简化复杂逻辑。
  • 避免大规模重写,聚焦小而安全的改进点。boy-scout-rule 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 建议结合自动化工具确保改动符合团队规范。

SKILL.md

Boy Scout Rule

"Leave the campground cleaner than you found it."

Always leave code better than you found it. Make incremental improvements when you touch a file.

What to Improve

Code Quality:

  • Remove dead code (commented blocks, unused functions)
  • Fix linting issues in files you touch
  • Improve unclear naming (x, temp, data → descriptive)
  • Add type annotations (TypeScript/Elixir @spec)
  • Extract magic numbers to named constants
  • Simplify complex logic
  • Add missing error handling
  • Update outdated comments
  • Fix formatting
  • Remove unused imports/variables
  • Consolidate duplication

What NOT to Do

  • ❌ Massive unrelated refactors
  • ❌ Change behavior without tests
  • ❌ Fix everything in the file (stay focused)
  • ❌ Breaking changes without tests
  • ❌ Premature optimization
  • ❌ Change unrelated sections

Process

  1. Before changes: Read file, note obvious issues, run linter
  2. Make primary changes: Implement feature/fix, write tests
  3. Apply improvements: Fix linting, improve naming, add types, extract constants, remove dead code
  4. Run verification: mix lint && mix test or yarn test:lint && yarn ts:check && yarn test
  5. Document: Include boy scout changes in commit message

Example Commit Message

Add worker search filter

- Implement location-based filtering
- Add tests for search radius

Boy Scout improvements:
- Extract SEARCH_RADIUS constant
- Add type annotations to helper functions
- Remove unused import statements
- Improve variable naming in search logic

Example Improvements

Before:

function calculateTotal(items) {  // No types
  let t = 0;  // Poor naming
  for (let i = 0; i < items.length; i++) {  // Old-style
    t += items[i].price * 1.08;  // Magic number
  }
  return t;
}

After:

const TAX_RATE = 1.08;

function calculateTotal(items: Item[]): number {
  return items.reduce((total, item) => {
    return total + (item.price * TAX_RATE);
  }, 0);
}

Improvements: types, constant, naming, modern syntax, simplified.

Critical Rules

  • Keep improvements in same commit as primary changes
  • Focus on "blast radius" (code near your changes)
  • Prioritize readability over cleverness
  • Always run full test suite after improvements
  • Ask for help when unsure about business logic
  • Small improvements > perfect refactors

Checklist

  • Removed dead code in files I touched
  • Fixed linting issues
  • Improved naming where I made changes
  • Added type annotations where missing
  • Extracted magic numbers
  • Updated outdated comments
  • Removed unused imports
  • Simplified complex logic
  • Added error handling
  • All tests pass
  • No new linting errors
  • Documented in commit message

Integration

During implementation: Make changes, apply boy scout, verify, commit together

During code review: Look for boy scout opportunities, recognize good boy scouting

During bug fixes: Fix bug, improve surrounding code, add tests, clean up

Remember

Incremental improvement, not perfection

  • Small improvements compound over time
  • Every file touched is an opportunity
  • Be a good steward of the codebase
  • When in doubt, ask for review on larger improvements

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

29.13%
按下载量换算54

Codex

22.67%
按下载量换算42

OpenCode

16.31%
按下载量换算30

Antigravity

12.75%
按下载量换算23

windsurf

7.52%
按下载量换算14

Gemini CLI

3.42%
按下载量换算6

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills