Token导航 LogoToken导航TokenDH.com
研究检索权限需确认github未标认证来源可访问clear审计未展示

laravel_prompt-structureLaravel prompt structure 搜索

Agent Skill

用于辅助提示词、系统指令、Agent 行为约束和工作流模板的整理。它适合让 Agent 规范任务边界、统一输出格式、拆分操作步骤或优化提示词可复用性。使用时需要保留真实业务约束,不要把示例当硬规则;涉及自动执行、外部工具或高风险操作时,应在提示词中明确确认步骤、权限边界和失败处理方式。

总安装

979

周安装

40

GitHub Stars

公开资料未说明

下载量

317
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

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:laravel_prompt-structure(Laravel prompt structure 搜索)
来源仓库:https://github.com/jpcaparas/superpowers-laravel
仓库路径:skills/laravel_prompt-structure
安装命令:
npx skills add jpcaparas/superpowers-laravel --skill "laravel:prompt-structure"
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。

AgentSkills.tonpx skills
npx skills add jpcaparas/superpowers-laravel --skill "laravel:prompt-structure"

简介

用于辅助提示词结构设计与 Agent 行为规范,提升任务边界清晰度与输出一致性。

  • 适用于 Codex、Claude、Cursor、Gemini CLI 等宿主环境中的提示工程与工作流优化场景。
  • 可帮助拆分操作步骤、统一输出格式,但需保留真实业务约束,勿将示例当作硬规则。
  • 安装命令为 npx skills add jpcaparas/superpowers-laravel --skill "laravel:prompt-structure"。
  • 涉及自动执行或外部工具时,应在提示词中明确确认步骤与权限边界。

SKILL.md

Prompt Structure

Well-structured prompts get complete, actionable responses. Poor structure leads to back-and-forth clarifications and incomplete solutions.

Separate Concerns

Ineffective

"Update the User model to add email verification and also create an API endpoint for password reset and add validation for the profile update form"

Effective

Break into focused requests:

  1. "Add email verification to User model using Laravel's MustVerifyEmail trait"
  2. "Create password reset API endpoint at POST /api/password/reset"
  3. "Add validation to ProfileUpdateRequest for name, email, and bio fields"

Each request has one clear goal. Easier to review, test, and iterate.

Prioritize and Sequence

Ineffective

"Need authentication, user dashboard, admin panel, and email notifications"

Effective

"Let's build authentication first:

  1. Set up Sanctum for API authentication
  2. Create login/register endpoints
  3. Add middleware to protect routes

Once auth works, we'll add the user dashboard, then admin features, then notifications."

Sequence matters. Build foundation first, then features that depend on it.

Specify Acceptance Criteria

Vague

"Make the checkout process work"

Specific

"Implement checkout process with these requirements:

  • User must be authenticated
  • Cart must have at least one item
  • Calculate total including tax (8.5%)
  • Create Order record with 'pending' status
  • Dispatch ProcessOrderJob to queue
  • Return order ID and confirmation message
  • Handle payment failures gracefully"

Clear criteria = clear implementation.

Break Down Work

Too Large

"Build a complete blog system with posts, comments, categories, tags, and search"

Right-Sized

"Let's start with core post functionality:

Phase 1: Basic Posts

  • Create Post model (title, content, published_at)
  • Add CRUD endpoints
  • Write feature tests for create/read/update/delete

Phase 2: Categories

  • Add Category model with belongsToMany relationship
  • Update post endpoints to handle categories
  • Test category assignment

Phase 3: Comments (and so on...)

Each phase is testable and deliverable."

Refactoring Requests

Unclear

"This code is messy, clean it up"

Clear

"Refactor UserController@store:

Current problem:

  • 80 lines in one method
  • Mixes validation, business logic, and email sending
  • Hard to test individual pieces

Desired improvements:

  • Extract validation to UserStoreRequest
  • Move user creation logic to UserService
  • Dispatch SendWelcomeEmail job instead of inline sending
  • Controller should orchestrate, not implement"

Explain the problem and the desired state.

Planning Features

Ineffective

"Add social login"

Effective

"Add GitHub OAuth login:

Acceptance criteria:

  • User clicks 'Login with GitHub' button
  • Redirects to GitHub OAuth
  • On callback, create or find user by github_id
  • Store github_token for API access
  • Redirect to dashboard with session

Testable increments:

  1. Set up Socialite with GitHub provider
  2. Add github_id and github_token columns to users
  3. Create OAuth controller with redirect and callback methods
  4. Write feature test for full OAuth flow
  5. Add 'Login with GitHub' button to login page"

Break it down. Make it testable.

Quick Reference

Structure your prompts:

  • One concern per request - Don't mix unrelated changes
  • Sequence logically - Build dependencies first
  • Define success - What does "done" look like?
  • Break down large work - Aim for testable increments
  • Explain refactoring - Current problem + desired state

Clear structure = clear results.

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

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

28.26%
按下载量换算90

Antigravity

21.87%
按下载量换算69

windsurf

15.8%
按下载量换算50

OpenCode

11.44%
按下载量换算36

Gemini CLI

7.9%
按下载量换算25

Codex

3.74%
按下载量换算12

安全审计

暂无安全审计结果可展示。

权限和风险

权限需确认

当前来源未能明确判断权限范围,默认进入异常复核队列。

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