Token导航 LogoToken导航

Agent Skill · 研究检索

faq-builder(常见问题解答生成器)

faq-builder 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。

faq-builder 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,权限为只读;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 8,745。

总安装 15,700周安装 710GitHub Stars 3下载 9,493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:faq-builder(常见问题解答生成器)
来源仓库:https://github.com/jmsktm/claude-settings
仓库路径:skills/faq-builder
安装命令:
npx skills add https://github.com/jmsktm/claude-settings --skill 'FAQ Builder'
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/jmsktm/claude-settings --skill 'FAQ Builder'

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、检索和筛选常见问题解答构建方法与模板。适合根据产品功能、用户痛点或知识盲点定位高频问题。使用时需提供完整功能说明、错误场景和预期行为。faq-builder 属于研究检索类 Skill,可作为该场景下的辅助能力补充。结果引用建议结合原始 README 核验具体用法和限制。

安装前建议确认权限范围及是否会触发联网或文件读写操作。

来源仓库内路径:skills/faq-builder

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/jmsktm/claude-settings --skill 'FAQ Builder';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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

faq-builder 怎么安装?

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

faq-builder 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。

faq-builder 需要什么权限?

权限级别是只读。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。

faq-builder 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。

faq-builder 用的人多吗?

公开数据是累计安装 15,700 次,近一周安装 710 次,下载 9,493 次,GitHub 3 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。

faq-builder 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。
  • 标注为只读权限,试用阶段不容易误改本地文件。

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。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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。
  • 只想先试试效果、不希望它改动本地文件的人。

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
faq-builder

常见问题解答生成器 · 本页

15,7003只读未明确有提醒

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

FAQ Builder

Expert FAQ database creation system that transforms support tickets, documentation, and user interactions into comprehensive, searchable FAQ content. This skill provides structured workflows for identifying common questions, crafting clear answers, organizing content hierarchically, and maintaining FAQ freshness over time.

FAQs are the foundation of customer self-service. Well-built FAQ databases deflect support tickets, reduce customer frustration, and free up support teams for complex issues. This skill helps you build FAQs that actually get used and answer questions customers actually ask.

Built on best practices from support operations and content strategy, this skill combines data mining, content writing, and continuous improvement to create FAQs that scale with your product.

Core Workflows

Workflow 1: Question Discovery

Mine support data to identify what customers actually ask

  1. Support Ticket Analysis

- Export recent tickets (90 days recommended) - Categorize by topic/feature area - Identify recurring questions (3+ occurrences) - Note exact phrasing customers use - Track question frequency and trends

  1. Search Query Mining

- Analyze help center search queries - Identify zero-result searches (questions without answers) - Track most common search terms - Note failed search patterns - Map search queries to existing content gaps

  1. Customer Interaction Sources

- Live chat transcripts - Community forum posts - Social media mentions - Sales call objections - Onboarding call questions - NPS/CSAT survey comments

  1. Internal Knowledge Sources

- Support team tribal knowledge - Sales FAQ documents - Product documentation gaps - Engineering known issues - Feature release notes

Workflow 2: FAQ Structure Design

Organize questions into logical, navigable hierarchy

  1. Category Architecture

- Getting Started: Account setup, first steps, basic concepts - Features & Usage: How to use specific functionality - Billing & Payments: Pricing, invoices, upgrades - Account Management: Settings, security, data - Troubleshooting: Common issues and fixes - Integrations: Third-party connections

  1. Question Hierarchy

- Top-level categories (5-8 maximum) - Subcategories (3-7 per category) - Individual questions (5-15 per subcategory) - Related questions linking

  1. Navigation Design

- Search functionality (primary) - Category browsing (secondary) - Popular questions section - Recently updated section - Contextual FAQs (in-app placement)

Workflow 3: Answer Crafting

Write clear, helpful answers that resolve questions

  1. Answer Structure Template [Direct Answer] - 1-2 sentences answering the question directly [Explanation] - Why/how, if needed for context [Steps] - Numbered steps if action required [Screenshot/Visual] - If helpful [Related Questions] - Links to related content [Still Need Help?] - Escalation path
  2. Writing Guidelines

- Lead with the answer (no preamble) - Use customer language, not internal jargon - Keep answers scannable (bullets, headers) - Include visuals for complex processes - Provide concrete examples - Link to detailed docs for deep dives

  1. Answer Quality Checklist

- Answers the question directly in first sentence - Uses language customers use - Includes necessary steps/actions - Has visual aid if process is complex - Links to related content - Provides escalation path - Tested by non-expert

Workflow 4: FAQ Maintenance

Keep content accurate and fresh

  1. Update Triggers

- Product feature changes - UI/UX updates - Pricing changes - Policy updates - New common questions emerge - Answer quality issues flagged

  1. Review Cadence

- Weekly: Check for outdated screenshots/steps - Monthly: Review question frequency data - Quarterly: Full content audit - On Release: Update affected FAQs

  1. Quality Monitoring

- Track "Was this helpful?" feedback - Monitor escalation rates from FAQs - Analyze search-to-ticket correlation - Review support ticket references to FAQs

  1. Retirement Process

- Archive deprecated feature FAQs - Redirect old URLs - Update cross-references - Remove from search index

Workflow 5: FAQ Analytics & Optimization

Measure effectiveness and improve continuously

  1. Key Metrics

- Deflection Rate: % of visitors who don't create tickets - Search Success Rate: % finding answers via search - Helpful Ratings: % marked as helpful - Time on Page: Engagement indicator - Exit to Ticket: FAQs that fail to resolve

  1. Optimization Actions

- Low helpful rating → Rewrite answer - High exit-to-ticket → Answer incomplete - Low views but high ticket volume → Discovery problem - Zero-result searches → Content gap - High search but low click → Title/description mismatch

  1. A/B Testing

- Test answer formats (video vs. text) - Test question phrasing - Test category organization - Test CTA placement

Quick Reference

ActionCommand/Trigger
Analyze tickets for FAQs"Mine tickets for FAQ opportunities"
Generate FAQ from topic"Create FAQ for [feature]"
Audit FAQ coverage"Audit FAQ completeness"
Review FAQ metrics"Show FAQ performance"
Find content gaps"Identify FAQ gaps"
Rewrite poor FAQ"Improve FAQ for [question]"
Structure FAQ hierarchy"Design FAQ categories"
Check FAQ freshness"Audit FAQ currency"
Generate FAQ from docs"Convert docs to FAQ"
Merge duplicate FAQs"Consolidate FAQ duplicates"

Best Practices

Question Writing

  • Use the exact words customers use
  • Start with action verbs: "How do I...", "What is...", "Can I..."
  • Be specific: "How do I reset my password?" not "Password help"
  • Match search intent: question should match what users type
  • Avoid internal terminology

Answer Writing

  • Answer in the first sentence - no lead-up
  • Use second person ("you") not passive voice
  • Break complex answers into numbered steps
  • Include screenshots for UI-related answers
  • Keep answers concise (target under 300 words)
  • Link to detailed docs for comprehensive coverage

Organization

  • Limit top-level categories to 5-8
  • Most popular questions should be 2 clicks maximum
  • Cross-link related questions
  • Use consistent naming conventions
  • Enable robust search functionality

Maintenance

  • Assign ownership for each FAQ section
  • Create alerts for product changes affecting FAQs
  • Track "Was this helpful?" religiously
  • Archive rather than delete old content
  • Keep version history

Integration

  • Embed contextual FAQs in product UI
  • Surface FAQs in chatbot/support flow
  • Include in onboarding sequences
  • Reference in support macros
  • Add to knowledge base search

FAQ Templates

Simple Question Template

## [Question in customer language]

[Direct answer in 1-2 sentences.]

[Additional context if needed.]

**Related questions:**
- [Related FAQ 1]
- [Related FAQ 2]

*Last updated: [Date]*

Process Question Template

## How do I [action]?

[Brief explanation of what this does.]

### Steps

1. [First step]
   ![Screenshot if helpful]
2. [Second step]
3. [Third step]

### Notes
- [Any important considerations]
- [Common gotchas]

**Need help?** [Contact support link]

*Last updated: [Date]*

Troubleshooting Template

## [Problem statement as customer would phrase it]

This usually happens when [common cause].

### Quick fixes

1. **Try this first**: [Most common solution]
2. **If that doesn't work**: [Second option]
3. **Still having issues**: [Third option]

### When to contact support
If none of the above work, [contact support] with:
- [Information to include]
- [Error messages/screenshots]

*Last updated: [Date]*

Red Flags

  • High exit-to-ticket rate: Answer isn't resolving the question
  • Low search success: Content doesn't match search terms
  • Outdated screenshots: Product has changed
  • Internal jargon: Customers can't understand
  • No escalation path: Dead ends frustrate users
  • Duplicate content: Confuses and dilutes SEO
  • Orphaned FAQs: No navigation path to content

Integration Points

  • Support Ticketing: Auto-suggest FAQs before ticket creation
  • Chatbot: FAQ as first-line response source
  • In-App Help: Contextual FAQ widgets
  • Search: Unified search across FAQ and docs
  • Analytics: Track deflection and engagement
  • CMS: Structured content management

平台分布

Codex

36.09%
按下载量换算3,426

Claude

29.81%
按下载量换算2,830

Cursor

20.96%
按下载量换算1,990

Gemini CLI

10.55%
按下载量换算1,002

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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