Token导航 LogoToken导航

Agent Skill · 研究检索

find-rules(寻找规则)

find-rules 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。

find-rules 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 32,976。

总安装 2,444周安装 105GitHub Stars 323下载 857核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:find-rules(寻找规则)
来源仓库:https://github.com/pedronauck/skills
仓库路径:skills/find-rules
安装命令:
npx skills add https://github.com/pedronauck/skills --skill find-rules
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/pedronauck/skills --skill find-rules

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。适用于关键词搜索、任务场景匹配或来源线索筛选等研究检索场景。通过 npx skills add 命令从指定 GitHub 仓库安装并使用。安装前需确认权限范围、维护状态及是否触发联网或文件操作。

find-rules 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/find-rules

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/pedronauck/skills --skill find-rules;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

find-rules 怎么安装?

在终端执行 npx skills add https://github.com/pedronauck/skills --skill find-rules。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

find-rules 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

find-rules 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

find-rules 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

find-rules 用的人多吗?

公开数据是累计安装 2,444 次,近一周安装 105 次,下载 857 次,GitHub 323 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

find-rules 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
find-rules

寻找规则 · 本页

2,444323需要联网未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Find Rules

This skill helps you discover and understand the specific rules, conventions, and guidelines that apply to your current task. It systematically explores project documentation to ensure compliance with project standards.

Instructions

Follow this workflow to identify applicable rules:

1. Understand Context

First, analyze the task to determine:

  • Task Type: PRD creation, Technical Spec, Feature Implementation, Refactoring, Bugfix, etc.
  • Technologies: React, Elysia, Tailwind, Drizzle, etc.
  • Aspects: Architecture, Testing, Styling, Naming, etc.

2. Pattern-Based Discovery

REQUIRED: Use Glob and Grep tools for local code discovery to find relevant rule files and guidelines.

EXECUTION STRATEGY: Execute 3+ tools simultaneously in the FIRST action for efficiency. Never execute tools sequentially when they can run in parallel.

MANDATORY STEPS:

  1. Find Rule Files (run in parallel): Use multiple Glob patterns simultaneously. # Primary rules Glob pattern=".cursor/rules/*.mdc" Glob pattern="**/CLAUDE.md" Glob pattern="**/AGENTS.md" # Secondary rules Glob pattern="**/CONTRIBUTING.md" Glob pattern="**/ARCHITECTURE.md" Glob pattern="**/STYLEGUIDE.md" # Alternative AI tool locations Glob pattern=".windsurf/rules/*" Glob pattern=".cursorrules" Glob pattern=".copilot/*" # Configuration files Glob pattern="**/.eslintrc*" Glob pattern="**/tsconfig.json" Glob pattern="**/biome.json"
  2. Search by Content: Use Grep to find rules mentioning specific technologies or concepts. Grep pattern="react" path=".cursor/rules" Grep pattern="CRITICAL|MANDATORY" path=".cursor/rules" output_mode="content" Grep pattern="testing" glob="*.md" Grep pattern="must|should|never" glob="*.mdc" -i=true
  3. Search Locations Priority:

- .cursor/rules/*.mdc - Technology-specific rules (highest priority) - CLAUDE.md - General project guidelines - AGENTS.md - Agent-specific instructions - .windsurf/rules/, .cursorrules - Alternative AI tool rules - docs/ - Documentation directory - Configuration files - Implicit rules from tooling

3. Extract and Categorize Rules

Once you identify the relevant files:

  1. Read: Use the Read tool to examine the full content of the identified rule files (e.g., .cursor/rules/react.mdc).
  2. Categorize Each Rule: For each rule found, determine:

- Category: Domain area (e.g., Authentication, Testing, Security, Performance, Styling, Architecture) - Content: The actual rule or guideline text - Source File: File path where the rule was found - Enforcement Level: CRITICAL, MANDATORY, RECOMMENDED, or SUGGESTED

  1. Analyze:

- Look for CRITICAL or MANDATORY requirements first - Identify naming conventions, file structures, and required patterns - Note any conflicts or specific instructions for the technologies involved - Check for keywords: "must", "should", "never", "always", "required", "forbidden"

4. Present Findings

Summarize the rules you found in a clear, actionable format:

  • Rule/Guideline: The specific standard.
  • Source: File path and section where it is defined.
  • Enforcement Level: CRITICAL, MANDATORY, RECOMMENDED.
  • Applicability: Why it applies to this task.

Documentation Sources

Primary Rule Locations

  • **.cursor/rules/*.mdc**: Technology-specific rules (React, Backend, Tailwind, etc.)
  • CLAUDE.md: General AI assistant guidelines and project context
  • AGENTS.md: Agent-specific instructions and workflows

Secondary Rule Locations

  • CONTRIBUTING.md: Contribution guidelines and processes
  • CODE_OF_CONDUCT.md: Community and code standards
  • ARCHITECTURE.md: Architectural decisions and patterns
  • STYLEGUIDE.md: Code style conventions
  • CODING_STANDARDS.md: Programming standards

Alternative AI Tool Locations

  • .windsurf/rules/: Windsurf-specific rules
  • .cursorrules: Cursor rules file
  • .copilot/: GitHub Copilot configuration
  • .codeium/: Codeium configuration

Configuration Files (Implicit Rules)

  • **.eslintrc.*,.eslintrc.json**: Linting rules
  • **.prettierrc.*,.prettierrc.json**: Formatting rules
  • tsconfig.json: TypeScript compiler options
  • .oxlintrc.json: Oxlint configuration
  • biome.json: Biome formatter/linter rules

Documentation Directories

  • docs/: Project documentation
  • .github/: GitHub-specific templates and workflows

Enforcement Levels

  • 🚨 CRITICAL: Must be followed strictly; violation leads to task rejection.
  • ⚠️ MANDATORY: Required standards that should not be skipped.
  • ✅ RECOMMENDED: Best practices that should be followed when applicable.
  • 💡 SUGGESTED: Nice-to-have guidelines.

Common Rule Categories

When categorizing rules, use these common domain areas:

  • Architecture: Design patterns, module boundaries, dependencies
  • Testing: Test requirements, coverage, patterns
  • Security: Authentication, authorization, input validation
  • Performance: Optimization requirements, caching, lazy loading
  • Styling: CSS/Tailwind conventions, component styling
  • Naming: File naming, variable naming, function naming
  • Code Quality: Linting, formatting, type safety
  • API: Endpoint patterns, request/response handling
  • State Management: Redux, Context, hooks patterns
  • Error Handling: Exception handling, error boundaries

Examples

User: "I need to create a new React component." Skill Action:

  1. Context: Feature development, React, component structure.
  2. Discovery:

- Use Glob: Glob pattern=".cursor/rules/*.mdc" to find rule files. - Use Grep: Grep pattern="react" path=".cursor/rules" to find React-related rules. - Result points to .cursor/rules/react.mdc.

  1. Extraction: Read .cursor/rules/react.mdc.
  2. Findings:

- 🚨 CRITICAL: Use functional components exclusively. - ⚠️ MANDATORY: File naming must be kebab-case.tsx. - ⚠️ MANDATORY: Use hooks for state management.

User: "How do I write a backend endpoint?" Skill Action:

  1. Context: Backend, API, implementation.
  2. Discovery:

- Use Glob: Glob pattern=".cursor/rules/*.mdc" to list available rules. - Use Grep: Grep pattern="backend|endpoint|elysia" path=".cursor/rules" to find backend rules. - Result points to .cursor/rules/elysia.mdc and .cursor/rules/data-fetch.mdc.

  1. Extraction: Read relevant files.
  2. Findings:

- 🚨 CRITICAL: Follow Elysia handler patterns. - ⚠️ MANDATORY: Use specific data fetching wrappers.

平台分布

Codex

37.66%
按下载量换算323

Claude

29.82%
按下载量换算256

Cursor

20%
按下载量换算171

Gemini CLI

9.5%
按下载量换算81

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills