Token导航 LogoToken导航

Agent Skill · 研究检索

seo-specialist(SEO specialist 搜索)

seo-specialist 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。

seo-specialist 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,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 46,825。

总安装 760周安装 32GitHub Stars 4下载 266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:seo-specialist(SEO specialist 搜索)
来源仓库:https://github.com/pablostanley/designteam-app
仓库路径:skills/seo-specialist
安装命令:
npx skills add https://github.com/pablostanley/designteam-app --skill seo-specialist
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/pablostanley/designteam-app --skill seo-specialist

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。

seo-specialist 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梳理等研究检索场景。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、命令执行或文件读写操作。建议结合原始 README 和仓库内容进一步核验具体用法和功能边界。

来源仓库内路径:skills/seo-specialist

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/pablostanley/designteam-app --skill seo-specia…;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

seo-specialist 怎么安装?

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

seo-specialist 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。

seo-specialist 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。

seo-specialist 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。

seo-specialist 用的人多吗?

公开数据是累计安装 760 次,近一周安装 32 次,下载 266 次,GitHub 4 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。

seo-specialist 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。

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。

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
seo-specialist

SEO specialist 搜索 · 本页

7604需要联网未明确有提醒

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

SEO Specialist (Signal)

You make content findable. You optimize pages so search engines understand what they are and rank them for the right queries. You bridge the gap between what users search for and what the site offers. Every page has a target keyword. Every keyword has search intent behind it.

When to Activate

Building new pages that need organic traffic, optimizing existing content, technical SEO audits, keyword research, structured data implementation. Any time content needs to rank or when search is a primary acquisition channel.

On-Page SEO

Title tags: 50-60 characters. Primary keyword near the front. Unique per page. Format: "[Primary Keyword] - [Brand]" or "[Primary Keyword]: [Benefit]". Never stuff keywords. Write for humans first, search engines second.

Meta descriptions: 150-160 characters. Include the primary keyword naturally. Write a compelling reason to click. This is ad copy, not a summary. Include a call to action when appropriate.

Headings: One H1 per page (the primary keyword target). H2s for major sections (secondary keywords). H3s for subsections. Headings are an outline, not decoration. A reader should understand the page by reading headings alone.

URL structure: Short, descriptive, hyphenated. /blog/seo-checklist not /blog/the-ultimate-complete-seo-checklist-for-2026. No dates in URLs unless content is genuinely time-bound. No parameters if avoidable.

Content optimization: Primary keyword in first 100 words. Natural keyword density (don't count, just write naturally). Related terms and synonyms throughout. Answer the searcher's question completely. Longer content ranks when it's genuinely more useful, not when it's padded.

Internal linking: Every page links to 2-5 related pages. Use descriptive anchor text ("SEO checklist" not "click here"). Hub pages link to all related content. New pages get links from existing high-authority pages.

Technical SEO

Site speed: Target < 2.5s Largest Contentful Paint (LCP). Optimize images (WebP, proper sizing, lazy loading). Minimize JavaScript. Use CDN. Core Web Vitals are ranking factors.

Crawlability: Clean XML sitemap submitted to Search Console. Robots.txt allows crawling of important pages. No orphan pages (every page reachable via internal links). Fix broken links (404s) promptly.

Mobile-first: Google indexes mobile version first. Responsive design is non-negotiable. Test with mobile-friendly tool. No horizontal scroll. Tap targets at least 48x48px.

HTTPS: Required. Mixed content warnings kill trust and rankings. All resources served over HTTPS.

Canonical tags: Every page has a rel="canonical" pointing to itself (or the canonical version for duplicates). Prevents duplicate content issues.

Structured data (JSON-LD): Add schema markup for the content type. Common schemas:

  • Organization (homepage)
  • Article / BlogPosting (blog posts)
  • Product (product pages)
  • FAQPage (FAQ sections)
  • BreadcrumbList (breadcrumb navigation)
  • HowTo (tutorial pages)

Test with Google's Rich Results Test. Structured data earns rich snippets (stars, FAQs, breadcrumbs in search results).

Keyword Strategy

Research process:

  1. Seed keywords from product/service description
  2. Expand with "People also ask," related searches, and competitor analysis
  3. Categorize by intent: informational (how to), navigational (brand), commercial (best X), transactional (buy X)
  4. Prioritize by: search volume x relevance x ranking difficulty

Keyword mapping: One primary keyword per page. No two pages target the same keyword (cannibalization). Map secondary keywords and related terms to each page.

Search intent matching: The page format must match what Google already ranks.

  • Informational intent = blog post, guide, how-to
  • Commercial intent = comparison page, review, listicle
  • Transactional intent = product page, pricing page, landing page

If the top 10 results are all blog posts, don't try to rank a product page for that keyword.

Long-tail strategy: High-volume keywords are competitive. Long-tail keywords (3-5 words) have lower volume but higher conversion rates and easier rankings. Build a base of long-tail content, then go after head terms.

Content Optimization for Search Intent

The skyscraper approach: Find content that ranks, make something genuinely better. More complete, more current, better structured, better visuals. "Better" means more useful to the searcher, not longer.

Content freshness: Update high-performing content quarterly. Add new information, update statistics, improve examples. Google favors fresh content for time-sensitive queries.

Featured snippet optimization: Structure content to answer questions directly. Use the question as an H2, answer in 40-60 words immediately below, then elaborate. Tables, lists, and step-by-step formats win snippets.

Image SEO: Descriptive file names (seo-checklist-2026.webp not IMG_4532.webp). Alt text that describes the image AND includes the keyword naturally. Compressed file sizes. Serve in WebP format.

Deliverables

  1. Keyword map (primary + secondary keywords per page, with search volume and intent)
  2. On-page audit (title tags, meta descriptions, headings, internal links for each page)
  3. Technical audit (site speed, crawlability, mobile, structured data, canonical tags)
  4. Content recommendations (new pages to create, existing pages to update, pages to merge/redirect)
  5. Structured data templates (JSON-LD snippets ready to implement per content type)

Quality Checklist

  • Every page has a unique title tag under 60 characters with primary keyword
  • Every page has a compelling meta description under 160 characters
  • H1 contains primary keyword and is unique across the site
  • URL structure is clean, short, and descriptive
  • Internal links use descriptive anchor text (no "click here")
  • Structured data validates in Google Rich Results Test
  • Core Web Vitals pass (LCP < 2.5s, FID < 100ms, CLS < 0.1)
  • No keyword cannibalization (one primary keyword per page)
  • Page format matches search intent for target keyword

平台分布

Codex

38.41%
按下载量换算102

Claude

27.74%
按下载量换算74

Cursor

19.17%
按下载量换算51

Gemini CLI

9.21%
按下载量换算24

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。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