Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

dhh-ruby-styleDHH Ruby style 搜索

Agent Skill

dhh-ruby-style 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

212

周安装

9

GitHub Stars

195

下载量

74
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

2

许可证

unknown

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:dhh-ruby-style(DHH Ruby style 搜索)
来源仓库:https://github.com/microck/ordinary-claude-skills
仓库路径:skills/dhh-ruby-style
安装命令:
npx skills add https://github.com/microck/ordinary-claude-skills --skill dhh-ruby-style
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。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/microck/ordinary-claude-skills --skill dhh-ruby-style

简介

dhh-ruby-style 强制执行 DHH 认可的 Ruby 语言风格与开发规范体系。

  • 规定控制器动作长度不超过五行,鼓励使用 before_action 减少重复代码。
  • 要求测试文件严格镜像 lib 目录结构,采用 _test.dart 后缀命名约定。
  • 提倡 immutable 数据模型与依赖注入,拒绝可变全局状态污染测试环境。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

DHH Ruby/Rails Style Guide

Write Ruby and Rails code following DHH's philosophy: clarity over cleverness, convention over configuration, developer happiness above all.

Quick Reference

Controller Actions

  • Only 7 REST actions: index, show, new, create, edit, update, destroy
  • New behavior? Create a new controller, not a custom action
  • Action length: 1-5 lines maximum
  • Empty actions are fine: Let Rails convention handle rendering
class MessagesController < ApplicationController
  before_action :set_message, only: %i[ show edit update destroy ]

  def index
    @messages = @room.messages.with_creator.last_page
    fresh_when @messages
  end

  def show
  end

  def create
    @message = @room.messages.create_with_attachment!(message_params)
    @message.broadcast_create
  end

  private
    def set_message
      @message = @room.messages.find(params[:id])
    end

    def message_params
      params.require(:message).permit(:body, :attachment)
    end
end

Private Method Indentation

Indent private methods one level under private keyword:

  private
    def set_message
      @message = Message.find(params[:id])
    end

    def message_params
      params.require(:message).permit(:body)
    end

Model Design (Fat Models)

Models own business logic, authorization, and broadcasting:

class Message < ApplicationRecord
  belongs_to :room
  belongs_to :creator, class_name: "User"
  has_many :mentions

  scope :with_creator, -> { includes(:creator) }
  scope :page_before, ->(cursor) { where("id < ?", cursor.id).order(id: :desc).limit(50) }

  def broadcast_create
    broadcast_append_to room, :messages, target: "messages"
  end

  def mentionees
    mentions.includes(:user).map(&:user)
  end
end

class User < ApplicationRecord
  def can_administer?(message)
    message.creator == self || admin?
  end
end

Current Attributes

Use Current for request context, never pass current_user everywhere:

class Current < ActiveSupport::CurrentAttributes
  attribute :user, :session
end

# Usage anywhere in app
Current.user.can_administer?(@message)

Ruby Syntax Preferences

# Symbol arrays with spaces inside brackets
before_action :set_message, only: %i[ show edit update destroy ]

# Modern hash syntax exclusively
params.require(:message).permit(:body, :attachment)

# Single-line blocks with braces
users.each { |user| user.notify }

# Ternaries for simple conditionals
@room.direct? ? @room.users : @message.mentionees

# Bang methods for fail-fast
@message = Message.create!(params)
@message.update!(message_params)

# Predicate methods with question marks
@room.direct?
user.can_administer?(@message)
@messages.any?

# Expression-less case for cleaner conditionals
case
when params[:before].present?
  @room.messages.page_before(params[:before])
when params[:after].present?
  @room.messages.page_after(params[:after])
else
  @room.messages.last_page
end

Naming Conventions

ElementConventionExample
Setter methodsset_ prefixset_message, set_room
Parameter methods{model}_paramsmessage_params
Association namesSemantic, not genericcreator not user
ScopesChainable, descriptivewith_creator, page_before
PredicatesEnd with ?direct?, can_administer?

Hotwire/Turbo Patterns

Broadcasting is model responsibility:

# In model
def broadcast_create
  broadcast_append_to room, :messages, target: "messages"
end

# In controller
@message.broadcast_replace_to @room, :messages,
  target: [ @message, :presentation ],
  partial: "messages/presentation",
  attributes: { maintain_scroll: true }

Error Handling

Rescue specific exceptions, fail fast with bang methods:

def create
  @message = @room.messages.create_with_attachment!(message_params)
  @message.broadcast_create
rescue ActiveRecord::RecordNotFound
  render action: :room_not_found
end

Architecture Preferences

TraditionalDHH Way
PostgreSQLSQLite (for single-tenant)
Redis + SidekiqSolid Queue
Redis cacheSolid Cache
KubernetesSingle Docker container
Service objectsFat models
Policy objects (Pundit)Authorization on User model
FactoryBotFixtures

Detailed References

For comprehensive patterns and examples, see:

  • references/patterns.md - Complete code patterns with explanations
  • references/resources.md - Links to source material and further reading

Philosophy Summary

  1. REST purity: 7 actions only; new controllers for variations
  2. Fat models: Authorization, broadcasting, business logic in models
  3. Thin controllers: 1-5 line actions; extract complexity
  4. Convention over configuration: Empty methods, implicit rendering
  5. Minimal abstractions: No service objects for simple cases
  6. Current attributes: Thread-local request context everywhere
  7. Hotwire-first: Model-level broadcasting, Turbo Streams, Stimulus
  8. Readable code: Semantic naming, small methods, no comments needed
  9. Pragmatic testing: System tests over unit tests, real integrations

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

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

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

Codex

38.08%
按下载量换算28

Claude

28.62%
按下载量换算21

Cursor

19.16%
按下载量换算14

Gemini CLI

9.08%
按下载量换算7

安全审计

Gen Agent Trust Hub

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills