Token导航 LogoToken导航

Agent Skill · 研究检索

featured-snippet(特色片段)

featured-snippet 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。

featured-snippet 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,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 10,839。

总安装 12,042周安装 512GitHub Stars 409下载 4,219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:featured-snippet(特色片段)
来源仓库:https://github.com/kostja94/marketing-skills
仓库路径:skills/featured-snippet
安装命令:
npx skills add https://github.com/kostja94/marketing-skills --skill featured-snippet
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/kostja94/marketing-skills --skill featured-snippet

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。

特色片段技能用于查找、检索和筛选相关信息。featured-snippet 属于研究检索类 Skill,可作为该场景下的辅助能力补充。适合根据关键词或任务场景快速定位候选结果。可结合来源仓库和原始 README 继续核验具体用法。安装前建议确认权限范围和维护状态,避免触发不必要的网络请求。

注意区分本地环境与生产环境的差异,防止误操作。

来源仓库内路径:skills/featured-snippet

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/kostja94/marketing-skills --skill featured-sni…;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

featured-snippet 怎么安装?

在终端执行 npx skills add https://github.com/kostja94/marketing-skills --skill featured-snippet。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

featured-snippet 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。

featured-snippet 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。

featured-snippet 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。

featured-snippet 用的人多吗?

公开数据是累计安装 12,042 次,近一周安装 512 次,下载 4,219 次,GitHub 409 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。

featured-snippet 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。
  • 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
featured-snippet

特色片段 · 本页

12,042409只读未明确通过

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

SEO On-Page: Featured Snippet

Guides optimization for Featured Snippets (Position Zero)—direct answers displayed above organic results. Featured snippets appear on ~19% of queries; by 2025, AI Overviews replaced many, but snippet optimization still supports AI citation and PAA. See serp-features for full SERP context.

When invoking: On first use, if helpful, open with 1–2 sentences on what this skill covers and why it matters, then provide the main output. On subsequent use or when the user asks to skip, go directly to the main output.

Scope

  • Featured Snippet formats: Paragraph, list, table
  • Content structure: Answer-first, H2/H3, semantic HTML
  • Query targeting: How, what, why; positions 2–5 opportunity
  • AI Overviews: Layered content for both snippets and AI citation

Current Landscape (2025–2026)

  • Featured snippets appear on ~19% of search queries
  • AI Overviews replaced 83%+ of featured snippets in many regions; appear in ~47% of US searches
  • When AI Overviews show, 83% of searches may end without clicks (zero-click)—goal shifts to citation as well as CTR. Featured Snippets also cause zero-click when the answer suffices. See serp-features for zero-click by feature.
  • Position zero still captures ~35% of clicks when snippets appear; pages with snippets see ~8% higher organic CTR
  • Optimize for both traditional snippets and AI Overviews

Featured Snippet vs AI Overview

Not the same. Featured snippets pull a direct passage from a single webpage. AI Overviews generate multi-source summaries using AI. When both appear, featured snippets still occupy a highly visible spot. Not every query triggers AI Overview—snippet optimization remains valuable. Semrush

Snippet Formats & Share

FormatShareUseOptimization
Paragraph~70%Definition, "what is," "why"40–60 words; direct answer after H2
List~19%How-to, steps, options<ol> or <ul>; semantic HTML
Table~6%Comparisons, stats, specsClear headers; target keywords in column/row headers
VideoRareVisual how-toVideo schema; timestamps/chapters; see video-optimization

45 words is the most common paragraph length. Answer-first format is critical. On mobile, featured snippets can occupy ~50% of the screen, pushing competitors below the fold. Semrush

Content Structure

Answer-First Approach

  • Place direct answer in first 1–2 sentences (40–60 words) before any other content
  • H2/H3: Use clear headings; Google extracts from content under headings
  • Semantic HTML: <ol>, <ul>, <table>, <p>—avoid divs styled as lists
  • Inverted pyramid: Answer immediately (for snippet/AI), then add nuance/data that earns clicks
  • Objective tone: Write like a dictionary entry—avoid personal opinions and emotional language; supports E-A-T

HTML Structure

ElementUse
H2/H3Question or topic; Google recognizes and extracts; list snippets can be compiled from headings across the page—Google pulls H2s and converts to list items
<ol>Steps, rankings, sequences; list snippets
<ul>Non-sequential items; bullet snippets
<table>Comparisons, specs; table snippets; clean rows/columns; H2/H3 intro; target keywords in heading and surrounding content; ensure data is relevant and scannable
First paragraph40–60 words; direct answer

Annotations matter: Search engines use semantic labeling to identify relevant blocks ("Fraggles") and extract passages. Structure pages clearly—headings, lists, tables—so crawlers can annotate and pull the right content. Annotations can suggest relationships between blocks, enabling engines to stitch text from multiple parts. SEJ / Bing

Visual Elements

  • Optimized images near answers can double CTR by capturing thumbnail slots
  • Alt text, captions, file names support Image Pack and snippet thumbnails. See image-optimization for captions and full image SEO; serp-features for Image Pack context.

Query Targeting

FactorGuidance
Query types"How," "what," "why" questions; definitional queries
Long-tailFeatured snippets typically appear for long-tail (specific) queries; short-tail is broad and less likely
Ranking positionTarget positions 2–5; snippet ownership often comes from these (not always #1). Prerequisite: Must rank in top ~20 blue links to be considered; Q&A runs through top organic results
Q&A memoryBing (and likely Google) memorize results; you may retain snippet even if blue link ranking drops—once earned, you don't necessarily need to maintain position
IntentInformational; match user intent precisely
ToolsPAA (excellent for low-volume, snippet-triggering queries), AnswerThePublic, AlsoAsked, Semrush Keyword Magic Tool (SERP features filter: "Featured snippet")

How the Algorithm Works (Bing/Google)

  • Foundation: Core algorithm ranks blue links; Q&A/Featured Snippet is a modular layer on top—same ranking foundation
  • Process: 1) Take top blue link results; 2) Extract or create summary; 3) Identify implicit question each document answers; 4) Match user's question to best implicit question; 5) Feature that answer
  • E-A-T: Snippets are E-A-T–based (Bing: "Relevancy" = accuracy/expertise). Correctness first—conforms to accepted opinion, document quality. Then authority and trust (document, author, publisher). See eeat-signals for E-E-A-T implementation.
  • Meta descriptions: Do not affect ranking; engines generate descriptions on the fly. Over-optimized or missing meta descriptions lead engines to extract their own—structure helps them choose well. SEJ

GEO / AI Overviews

  • Layered content: Clear definitions, lists, steps—AI can cite and summarize
  • Self-contained blocks: Each answer readable alone; supports AI extraction
  • Schema: FAQPage, HowTo, Article—helps machines understand structure
  • See generative-engine-optimization for full GEO strategy

Finding Opportunities

  • Manual: Search target keyword; note if snippet appears and format (paragraph/list/table/video)
  • PAA: People Also Ask questions often trigger snippets; use as H2 or dedicated page
  • Tools: Semrush Keyword Magic Tool—filter by "Featured snippet" in SERP features; "Open SERP" to view current snippet
  • Best opportunities: Keywords where you already rank on page 1 (positions 1–10)

Monitoring

  • Google Search Console: Identify snippet opportunities; track performance
  • Position Tracking (e.g. Semrush): "Featured Snippets" tab—"Already featured" vs "Opportunities"
  • Rich Results Test: Validate schema; ensure extraction eligibility
  • CTR benchmarks: Snippets ~114% CTR increase vs position #1; adjust expectations when AI Overviews dominate

Output Format

  • Target queries (snippet-worthy)
  • Format (paragraph/list/table) per query
  • Content structure (H2, answer length, semantic HTML)
  • Schema if applicable (FAQ, HowTo)

Related Skills

  • serp-features: SERP features overview; Featured Snippet in context
  • content-optimization: H2 keywords, tables, lists; defers snippet-specific structure to this skill
  • faq-page-generator: FAQ format and answer length; FAQ schema for PAA
  • howto-section-generator: HowTo sections; list snippets; step copy; HowTo schema alignment
  • schema-markup: FAQPage, HowTo for rich results
  • heading-structure: H2–H3 hierarchy
  • generative-engine-optimization: AI citation; GEO strategy
  • eeat-signals: E-E-A-T implementation; authority, trust, author
  • image-optimization: Image SEO for snippet thumbnails; alt, captions, file names; captions support snippet context
  • video-optimization: Video snippet format; VideoObject; timestamps/chapters

References

平台分布

Codex

34.42%
按下载量换算1,452

Claude

30.58%
按下载量换算1,290

Cursor

21.34%
按下载量换算900

Gemini CLI

8.97%
按下载量换算378

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。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