Token导航 LogoToken导航

Agent Skill · 研究检索

legal(legal 问题管理)

legal 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。

legal 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 56,831。

总安装 445周安装 18GitHub Stars 20下载 140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:legal(legal 问题管理)
来源仓库:https://github.com/howells/arc
仓库路径:skills/legal
安装命令:
npx skills add https://github.com/howells/arc --skill legal
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/howells/arc --skill legal

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,支持根据关键词或任务场景定位内容。适合在需要信息聚合的场景下使用,可结合来源仓库和原始 README 核验具体用法。安装命令:npx skills add https://github.com/howells/arc --skill legal。

适用于 Codex、Claude、Cursor、Gemini CLI,通过 GitHub 安装。建议确认权限范围和维护状态,注意是否触发联网或文件读写操作。

来源仓库内路径:skills/legal

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/howells/arc --skill legal;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

legal 怎么安装?

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

legal 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。

legal 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。

legal 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。

legal 用的人多吗?

公开数据是累计安装 445 次,近一周安装 18 次,下载 140 次,GitHub 20 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。

legal 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
legal

legal 问题管理 · 本页

44520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

<tool_restrictions>

MANDATORY Tool Restrictions

BANNED TOOLS — calling these is a skill violation:

  • EnterPlanMode — BANNED. Do NOT call this tool. This skill has its own multi-step process. Execute the steps below directly.
  • ExitPlanMode — BANNED. You are never in plan mode. </tool_restrictions>

<progress_context> Use Read tool: docs/arc/progress.md (first 50 lines)

Check for recent feature work that might affect data collection scope. </progress_context>

Legal Pages Workflow

Generate comprehensive legal pages (Privacy Policy, Terms of Service, Cookie Policy) through a guided, interactive process. Combines automatic project detection with user questions to create tailored documents.

These are starting points that MUST be reviewed by a qualified lawyer before publishing.


Process Overview

Step 1: Disclaimer & Scope     → Set expectations
Step 2: Project Detection      → Scan codebase for data collection
Step 3: Guided Questions       → Interactive Q&A to fill gaps (5 rounds)
Step 4: Generate Documents     → Create tailored legal pages
Step 5: Implementation         → Add to project with proper routing
Step 6: Next Steps            → Cookie consent, lawyer review, etc.

Step 1: Disclaimer & Scope

Always start with this disclaimer:

⚠️ Important: These are template documents, not legal advice. I'll generate comprehensive legal pages based on your project and answers, but: - I am not a lawyer and this is not legal advice - These templates should be reviewed by a qualified attorney - Laws vary by jurisdiction and change frequently - Regulated industries (healthcare, finance, children) have special requirements These documents will give you a solid starting point that covers common requirements under GDPR, CCPA, and general best practices.

Ask: "Do you want to proceed with generating legal pages for this project?"


Step 2: Project Detection

Perform comprehensive codebase scan for data collection signals.

Detection Checklist

Search for and report on:

AUTHENTICATION
├── next-auth / NextAuth.js    → OAuth providers, session strategy
├── clerk                       → User profiles, organizations
├── supabase auth              → Email, OAuth, phone auth
├── firebase auth              → Multiple auth methods
├── lucia                       → Session-based auth
├── auth0                       → Enterprise SSO, social login
├── passport.js                → Strategy-based auth
└── Custom auth                → JWT, session cookies

ANALYTICS & TRACKING
├── Google Analytics (gtag, GA4)
│   └── Cookies: _ga (2 years), _gid (24h), _gat (1 min)
├── Google Tag Manager         → Container for multiple tags
├── Plausible                  → Privacy-focused, no cookies
├── Fathom                     → Privacy-focused, no cookies
├── PostHog                    → Product analytics, session recording
├── Mixpanel                   → Event tracking, user profiles
├── Amplitude                  → Product analytics
├── Heap                       → Auto-capture analytics
├── Hotjar/FullStory          → Session recording, heatmaps
├── Vercel Analytics          → Privacy-focused, no cookies
└── Segment                    → Customer data platform

PAYMENTS & BILLING
├── Stripe
│   └── You store: customer_id, subscription status
│   └── Stripe stores: payment methods, card details
│   └── Cookies: __stripe_mid, __stripe_sid
├── Paddle                     → Merchant of record model
├── LemonSqueezy              → Merchant of record model
├── PayPal                     → Payment processor
└── Custom billing             → Invoice data, payment history

EMAIL SERVICES
├── Resend                     → Transactional email
├── SendGrid                   → Email delivery
├── Postmark                   → Transactional email
├── Mailchimp/ConvertKit      → Marketing email, subscriber lists
├── Customer.io               → Marketing automation
└── AWS SES                    → Email infrastructure

ERROR TRACKING & MONITORING
├── Sentry                     → Error tracking, may capture user context
├── LogRocket                  → Session replay, error tracking
├── Bugsnag                    → Error monitoring
├── Datadog                    → APM, logging, traces
└── New Relic                  → Application monitoring

CUSTOMER SUPPORT
├── Intercom                   → Chat, user data, conversation history
├── Crisp                      → Live chat
├── Zendesk                    → Support tickets
├── HelpScout                  → Customer support
└── Freshdesk                  → Support platform

DATABASE & STORAGE
├── PostgreSQL/MySQL          → User data storage
├── MongoDB                    → Document storage
├── Prisma                     → ORM (check schema for PII)
├── Drizzle                    → ORM
├── Supabase                   → Database + auth + storage
├── PlanetScale               → MySQL platform
├── Neon                       → Serverless Postgres
├── Cloudinary                → Image/video storage
├── Uploadthing               → File uploads
├── AWS S3                     → Object storage
└── Vercel Blob               → File storage

HOSTING & INFRASTRUCTURE
├── Vercel                     → Logs IP addresses, request data
├── Netlify                    → Similar logging
├── AWS                        → CloudFront logs, ALB logs
├── Cloudflare                → CDN, may set cookies
└── Railway/Render            → Platform logs

MARKETING & ADS
├── Facebook Pixel            → Conversion tracking
│   └── Cookies: _fbp, fr
├── Google Ads                → Conversion tracking
│   └── Cookies: _gcl_au, _gcl_aw
├── LinkedIn Insight Tag      → B2B tracking
├── Twitter/X Pixel           → Conversion tracking
├── TikTok Pixel              → Conversion tracking
└── Pinterest Tag             → Conversion tracking

CMS & CONTENT
├── Sanity                     → Content management
├── Contentful                → Headless CMS
├── Payload                    → Headless CMS
├── Strapi                     → Headless CMS
└── WordPress API             → Content source

FORMS & DATA COLLECTION
├── Contact forms              → Name, email, message
├── Newsletter signup         → Email address
├── User profiles             → Various PII
├── File uploads              → User-generated content
├── Surveys/feedback          → User responses
└── Job applications          → Resumes, personal info

Detection Output Format

Present findings to user:

## 📊 Data Collection Detection Results

### Authentication
**Detected:** NextAuth.js with Google and GitHub OAuth
- **Data collected:** Email, name, profile picture from OAuth providers
- **Data stored:** User record in database, session cookie
- **Session strategy:** JWT / Database sessions

### Analytics
**Detected:** Google Analytics 4
- **Data collected:** Page views, events, device info, IP address
- **Cookies set:**
  | Cookie | Purpose | Duration | Type |
  |--------|---------|----------|------|
  | _ga | Distinguishes users | 2 years | Analytics |
  | _gid | Distinguishes users | 24 hours | Analytics |

### Payments
**Detected:** Stripe
- **Data you store:** Customer ID, subscription status, billing address
- **Data Stripe stores:** Payment methods, transaction history
- **Note:** You are NOT a data controller for card numbers—Stripe is

### Third-Party Processors
| Service | Data Shared | Purpose | Their Privacy Policy |
|---------|-------------|---------|---------------------|
| Vercel | IP, request logs | Hosting | vercel.com/legal/privacy-policy |
| Resend | Email addresses | Transactional email | resend.com/legal/privacy-policy |
| Sentry | Error data, user context | Error tracking | sentry.io/privacy |

### Cookies Summary
| Category | Count | Examples |
|----------|-------|----------|
| Essential | 2 | Session, CSRF token |
| Analytics | 2 | _ga, _gid |
| Marketing | 0 | None detected |
| Functional | 1 | Theme preference |

Step 3: Guided Questions

Use AskUserQuestion tool for each round. One focused topic at a time.

Round 1: Business Identity

Question: "What are your business details?"
Header: "Business"
Options: [Free text response needed]

Gather:
- Legal business name (e.g., "Acme Inc." or "John Smith trading as Acme")
- Country/state of incorporation or residence
- Business type: Company, LLC, Sole proprietor, etc.
- Website URL
- Contact email for privacy/legal inquiries
- Physical address (required for some jurisdictions, recommended for all)

Round 2: Target Audience & Jurisdiction

Question: "Where are your users located?"
Header: "Jurisdiction"
Options:
  - "Worldwide (GDPR + CCPA compliant)" [Recommended]
    → Covers EU, California, and general best practices
  - "US only"
    → CCPA for California, general US practices
  - "EU/EEA only"
    → GDPR-focused
  - "Specific countries"
    → Ask follow-up for which countries

Follow-up if needed:
Question: "Do you expect users under 18?"
Header: "Age"
Options:
  - "No, adults only (18+)"
  - "Yes, 13-17 with parental consent"
  - "Yes, under 13" → COPPA applies, special handling required
  - "Not sure"

Round 3: Documents Needed

Question: "Which legal documents do you need?"
Header: "Documents"
MultiSelect: true
Options:
  - "Privacy Policy" [Required for almost all sites]
    → Required if you collect ANY data (even just analytics)
  - "Terms of Service"
    → Required for apps/SaaS, recommended for all
  - "Cookie Policy"
    → Required if using non-essential cookies (can be section in Privacy Policy)
  - "Acceptable Use Policy"
    → Recommended if users can post content or interact

Round 4: Service Type & Features

Question: "What type of service is this?"
Header: "Service type"
Options:
  - "SaaS / Web application"
    → User accounts, possibly subscriptions
  - "E-commerce / Online store"
    → Products, checkout, shipping
  - "Content / Blog / Marketing site"
    → Minimal data collection
  - "Marketplace / Platform"
    → Multiple user types, transactions between users
  - "API / Developer tools"
    → API keys, usage data, developer accounts

Follow-up based on selection:
- SaaS: "Do you offer free trials? Refund policy? Subscription billing?"
- E-commerce: "Physical or digital products? Return policy? Shipping regions?"
- Marketplace: "Do you facilitate payments between users? Take commission?"

Round 5: Specific Policies

Question: "What are your data practices?"
Header: "Practices"
MultiSelect: true
Options:
  - "We use data only for providing our service"
  - "We send marketing emails (with consent)"
  - "We share anonymized/aggregated data"
  - "We use AI/ML to process user data"
  - "We allow third-party integrations"

Question: "What is your refund/cancellation policy?"
Header: "Refunds"
Options:
  - "14-day money-back guarantee"
  - "30-day money-back guarantee"
  - "Pro-rated refunds for annual plans"
  - "No refunds (for digital goods)"
  - "Custom policy" → Ask for details

Step 4: Generate Documents

Based on detection + user answers, generate fully personalized documents.

CRITICAL: No Placeholders

DO NOT generate documents with [PLACEHOLDER] markers. The documents must be:

  • Filled in with actual company name, URLs, emails from user answers
  • Populated with actual detected services (Stripe, Vercel, etc.) by name
  • Include real cookie names and durations from detection
  • Have actual data categories based on what was detected
  • Remove sections that don't apply (e.g., no Payments section if no payments detected)

Example — WRONG:

We share data with [SERVICE_PROVIDERS].
Contact us at [EMAIL].

Example — CORRECT:

We share data with Vercel (hosting), Stripe (payments), and Resend (email).
Contact us at privacy@acme.com.

The templates below show the structure. When generating, replace ALL bracketed items with real values from detection and user answers. If a section doesn't apply to this project, omit it entirely.


Structure Reference Templates

The template structures for each document are maintained in separate files. Read these templates at runtime to use as the structure reference when generating personalized documents:

  • Privacy Policy: templates/privacy-policy.md
  • Terms of Service: templates/terms-of-service.md
  • Cookie Policy: templates/cookie-policy.md

Use Read tool to load each template before generating. The templates show the structure — when generating, replace ALL bracketed items with real values from detection and user answers. If a section doesn't apply to this project, omit it entirely.


Step 5: Implementation

Create the Pages

For Next.js App Router:

app/
├── (legal)/
│   ├── layout.tsx          # Shared layout for legal pages
│   ├── privacy/page.tsx    # Privacy Policy
│   ├── terms/page.tsx      # Terms of Service
│   └── cookies/page.tsx    # Cookie Policy (or section in privacy)

Example layout:

// app/(legal)/layout.tsx
export default function LegalLayout({ children }: { children: React.ReactNode }) {
  return (
    <div className="container mx-auto max-w-3xl px-4 py-12">
      <article className="prose prose-gray dark:prose-invert max-w-none">
        {children}
      </article>
    </div>
  )
}

Offer to create:

  1. The page files with generated content
  2. Footer links to the legal pages
  3. Cookie consent banner component (if needed)

Step 6: Next Steps

Present to user after generation:

## ✅ Legal Pages Generated

**Created:**
- `/privacy` — Privacy Policy
- `/terms` — Terms of Service
- `/cookies` — Cookie Policy

## Required Next Steps

1. **Add footer links**
   - Link to Privacy Policy, Terms, and Cookies from your site footer

2. **Cookie consent banner** (if using non-essential cookies)
   - Required before setting analytics/marketing cookies
   - Must offer "Reject All" option for GDPR compliance
   - Consider: [CookieConsent](https://github.com/orestbida/cookieconsent), [Osano](https://www.osano.com/), or custom

3. **Legal review**
   - Have these documents reviewed by a lawyer, especially if:
     - You handle sensitive data (health, financial)
     - You have users in multiple jurisdictions
     - You're in a regulated industry
     - You process children's data

4. **Keep updated**
   - Update "Last updated" date when you make changes
   - Review annually at minimum
   - Update when you add new data collection or third-party services

5. **Data Subject Requests**
   - Set up a process to handle privacy requests (access, deletion, etc.)
   - Aim to respond within 30 days (GDPR requirement)

<arc_log> After completing this skill, append to the activity log. See: references/arc-log.md

Entry: /arc:legal — Generated Privacy Policy, Terms, Cookie Policy </arc_log>


Interop

  • Invoked by /arc:letsgo when legal documents are missing
  • May invoke cookie consent implementation after generating Cookie Policy
  • References project detection patterns shared with /arc:letsgo

平台分布

Codex

37.87%
按下载量换算53

Claude

27.51%
按下载量换算39

Cursor

18.21%
按下载量换算25

Gemini CLI

9.18%
按下载量换算13

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