Token导航 LogoToken导航

Agent Skill · 研究检索

serp-feature-sniper(serp 功能狙击手)

serp-feature-sniper 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。

serp-feature-sniper 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,权限为接触敏感数据;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 70,176。

总安装 212周安装 9GitHub Stars 607下载 74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:serp-feature-sniper(serp 功能狙击手)
来源仓库:https://github.com/athina-ai/goose-skills
仓库路径:skills/serp-feature-sniper
安装命令:
npx skills add https://github.com/athina-ai/goose-skills --skill serp-feature-sniper
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/athina-ai/goose-skills --skill serp-feature-sniper

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。

注意该技能属于研究检索类别,适用于 Codex、Claude、Cursor、Gemini CLI。

来源仓库内路径:skills/serp-feature-sniper

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/athina-ai/goose-skills --skill serp-feature-sn…;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

    它需要接触敏感数据,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

serp-feature-sniper 怎么安装?

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

serp-feature-sniper 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。

serp-feature-sniper 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

serp-feature-sniper 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。

serp-feature-sniper 用的人多吗?

公开数据是累计安装 212 次,近一周安装 9 次,下载 74 次,GitHub 607 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。

serp-feature-sniper 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

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 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

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 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
serp-feature-sniper

serp 功能狙击手 · 本页

212607接触敏感数据未明确有提醒

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

SERP Feature Sniper

Most SEO content targets "page 1" generically. But the real estate that drives clicks — featured snippets, People Also Ask boxes, video carousels, image packs — follows specific formatting rules. This skill analyzes what SERP features exist for your keywords, who holds them, and builds content structures optimized to take them.

Core principle: A featured snippet isn't won by accident. It's won by formatting a specific answer in a specific way (paragraph, list, table) that matches what Google expects. This skill reverse-engineers the format Google wants, then tells your writer exactly how to structure their content.

When to Use

  • "How do I win featured snippets for these keywords?"
  • "What SERP features can we target?"
  • "Optimize our content for PAA boxes"
  • "Which of our pages could win SERP features with small changes?"
  • "SERP feature audit for our keyword list"

Tool Enhancement (Optional)

SERP feature analysis is dramatically better with real-time SERP data showing exactly which features appear, who holds them, and what content format they use.

Agent Prompt to User

"I can analyze SERP features and build content structures to win them. For the most accurate results — seeing exactly which features appear on each SERP and who currently holds them — I'd recommend connecting a SERP data API." Recommended: SerpAPI (100 free searches/month, $50/month for 5,000) - Sign up at serpapi.com → get API key - Set SERPAPI_KEY env var Alternatives that also work: - Serper.dev (2,500 free searches, then $50/month for 50K) → set SERPER_API_KEY - DataForSEO SERP (pay-per-use, ~$0.01/search) → set DATAFORSEO_LOGIN + DATAFORSEO_PASSWORD - ValueSERP (free tier available) → set VALUESERP_API_KEY "Want to use one of these, or should I proceed with baseline mode? Baseline uses web search to identify SERP features — still produces strong optimization plans, but with less precise feature detection and holder identification."

Mode Selection

  • Enhanced mode — Real SERP API data per keyword: exact features present, current feature holders, snippet content/format, PAA questions and their current answers, related searches. Enables precise "here's exactly what to write to steal this snippet" guidance.
  • Baseline mode — Uses web_search to analyze SERPs, fetch_webpage to study current feature holders. Feature detection is based on search result patterns. Still produces excellent optimization plans — the content structuring and formatting guidance doesn't change.

Phase 0: Intake

  1. Keywords — List of target keywords (1-100)
  2. Your site URL — So we can check if you already hold any features
  3. Existing content URLs — Pages you want to optimize for features (optional — if blank, we focus on new content)
  4. Priority features — Any specific features you care about? (featured snippets, PAA, video, images, all?)
  5. Content format flexibility — Can you add tables, lists, images, videos, or are you limited to text?
  6. Tool preference — Enhanced mode with SERP API, or baseline? (see Tool Enhancement above)

Phase 1: SERP Feature Audit

Enhanced Mode (SerpAPI / Serper / DataForSEO)

For each keyword, query the SERP API:

# SerpAPI example
GET https://serpapi.com/search?q={keyword}&api_key={key}&num=10

Extract per keyword:

  • Featured snippet — Present? Type (paragraph/list/table)? Current holder? Snippet content?
  • People Also Ask — Questions listed? How many? Current answers?
  • Video carousel — Present? Sources (YouTube, other)?
  • Image pack — Present? Source pages? Alt text patterns?
  • Knowledge panel — Present? Source? Editable?
  • Local pack — Present? (skip if not relevant)
  • Shopping results — Present? (skip if not relevant)
  • Top stories — Present? Sources?
  • Site links — Who has them?
  • FAQ rich results — Who has them?

Baseline Mode

For each keyword, use web_search:

  • Analyze result format for feature indicators
  • Check for "Featured snippet from..." patterns
  • Note PAA boxes that appear
  • Spot-check 3-5 results via fetch_webpage for schema markup

Phase 2: Feature Holder Analysis

For each keyword with winnable features, analyze the current holder:

2A: Featured Snippet Analysis

Fetch the current snippet holder's page via fetch_webpage:

  • Content format — Paragraph (40-60 words), numbered list, bulleted list, or table?
  • Placement on page — Where is the snippet content within the page structure?
  • Heading structure — What H2/H3 triggers the snippet?
  • Answer directness — Does the answer start immediately after the heading?
  • Page authority — Domain rating, page age
  • Vulnerabilities — Is the answer outdated? Too long? Missing context? Poorly formatted?

2B: PAA Analysis

For each PAA question:

  • Current answer source — Which page answers it?
  • Answer format — Paragraph, list, or table?
  • Answer length — How many words?
  • Answer quality — Comprehensive or thin?
  • Your coverage — Do you already answer this on your site?

2C: Rich Result Analysis

Check current top 5 pages for:

  • FAQ schema — Present? How many Q&A pairs?
  • HowTo schema — Present? Step count?
  • Review schema — Present? Rating?
  • Article schema — Present? Type?
  • Breadcrumb schema — Present?

Phase 3: Winnability Assessment

Score each keyword × feature combination:

FactorWeightDescription
Feature existsRequiredNo feature = nothing to win
Current holder weakness30%Outdated, thin, poorly formatted = easier to steal
Your domain authority25%Need minimum authority to compete
Content format match20%Can you produce the format Google wants?
Effort to create/modify15%New content vs. quick edit to existing page
Traffic value10%Search volume × CTR boost from feature

Winnability tiers:

  • High (70-100): Weak current holder + you have authority + low effort → prioritize
  • Medium (40-69): Competitive but achievable → include in content plan
  • Low (0-39): Strong holder or feature doesn't exist → deprioritize

Phase 4: Optimization Blueprints

For each winnable feature, produce specific formatting guidance:

Featured Snippet Blueprint

## Keyword: [target keyword]
**Current holder:** [URL] | **Format:** [paragraph/list/table]
**Winnability:** [High/Medium/Low]

### How to Win This Snippet

**Target heading:** Use this exact H2:
> ## [Heading that triggers the snippet — matches query intent]

**Answer format:** [Paragraph / Numbered list / Bulleted list / Table]

**If paragraph snippet:**
- Write a 40-60 word direct answer immediately after the H2
- Start with "[Keyword] is..." or "The [keyword] refers to..."
- Include the target keyword in the first sentence
- Be more precise and current than the current holder

**If list snippet:**
- Use a numbered or bulleted list of 5-8 items
- Each item should be 10-20 words
- Start each item with a strong verb or noun
- Include the complete list (not truncated)

**If table snippet:**
- Use a clean HTML/markdown table with 3-5 columns
- Include column headers that match the query dimensions
- Fill all cells with actual data (no blanks)
- Keep to 4-8 rows

**Content to write:**
[Specific paragraph/list/table draft optimized for snippet capture]

**Page placement:** Place this within the top 30% of page content

PAA Optimization Blueprint

## PAA Questions to Target

### Q: [Question 1]
**Current answer:** [summary] from [source]
**Your optimized answer:**
- H2: "[Question phrased as heading]"
- Answer: [50-80 word direct answer]
- Format: [paragraph/list]
- Include: [specific details to be more comprehensive than current answer]

### Q: [Question 2]
...

Schema Markup Blueprint

## Recommended Schema

### FAQ Schema
Add to page: [URL]

{ "@type": "FAQPage", "mainEntity": [ { "@type": "Question", "name": "[Question]", "acceptedAnswer": { "@type": "Answer", "text": "[Answer]" } } ] }


### HowTo Schema (if applicable)

[Step-by-step schema with estimated time and tools]

Phase 5: Output

# SERP Feature Sniper Report — [Site/Client] — [DATE]

## Executive Summary
- Keywords analyzed: [N]
- Keywords with winnable features: [N] ([%])
- High-priority targets: [N] featured snippets, [N] PAA boxes, [N] rich results
- Estimated CTR uplift if features won: [X]% average

---

## Feature Landscape

| Keyword | Volume | Featured Snippet | PAA | Video | Images | Rich Results | Your Status |
|---------|--------|-----------------|-----|-------|--------|-------------|-------------|
| [kw1] | [vol] | ✓ Paragraph — [holder] | ✓ 4 Qs | ✗ | ✓ | FAQ | Not ranking |
| [kw2] | [vol] | ✓ List — [holder] | ✓ 6 Qs | ✓ | ✗ | None | #5 |
| [kw3] | [vol] | ✗ | ✓ 3 Qs | ✗ | ✗ | HowTo | #3 |

---

## Priority Actions

### Tier 1: Quick Wins (existing pages, minor edits)
| Page | Keyword | Feature | Action Required | Effort |
|------|---------|---------|----------------|--------|
| [URL] | [kw] | Featured snippet | Add [format] answer under H2 "[heading]" | 30 min |
| [URL] | [kw] | FAQ rich result | Add FAQ schema with [N] Q&As | 15 min |

### Tier 2: Content Updates (existing pages, significant edits)
| Page | Keyword | Feature | Action Required | Effort |
|------|---------|---------|----------------|--------|
| [URL] | [kw] | Featured snippet | Restructure [section] as [format] | 2 hours |

### Tier 3: New Content Needed
| Keyword | Feature | Content Type | Blueprint Reference |
|---------|---------|-------------|-------------------|
| [kw] | Featured snippet + PAA | [type] | See Brief #[N] |

---

## Detailed Blueprints

### Blueprint 1: [Keyword]
[Full optimization blueprint per Phase 4]

### Blueprint 2: [Keyword]
...

---

## Schema Markup Checklist
- [ ] Add FAQ schema to [URL] — [N] Q&A pairs
- [ ] Add HowTo schema to [URL] — [N] steps
- [ ] Add Article schema to [URL] — [type]
- [ ] Validate all schema at search.google.com/test/rich-results

---

## Monitoring Plan
- Re-check SERP features monthly for target keywords
- Track feature wins/losses in [tool]
- Re-optimize if feature is lost or format changes

Save to clients/<client-name>/seo/serp-feature-report-[YYYY-MM-DD].md.

Cost

ComponentCost
SerpAPI (enhanced, per keyword)~$0.01-0.05 per keyword
Serper.dev (enhanced, per keyword)~$0.001 per keyword
Page fetches (feature holders)~$0.01-0.03 per keyword
SEO domain analyzer~$0.10-0.20
AnalysisFree (LLM reasoning)
Total per keyword (baseline)~$0.03-0.08
Total per keyword (enhanced)~$0.05-0.12
Batch of 50 keywords (baseline)~$1.50-4.00
Batch of 50 keywords (enhanced)~$2.50-6.00

Tools Required

  • Upstream skills: seo-domain-analyzer, fetch_webpage, web_search
  • Optional (enhanced): SerpAPI (SERPAPI_KEY), Serper.dev (SERPER_API_KEY), DataForSEO (DATAFORSEO_LOGIN + DATAFORSEO_PASSWORD), or ValueSERP (VALUESERP_API_KEY)

Scheduling

For ongoing SERP feature monitoring:

  • Run monthly against target keyword list
  • Compare with previous month's report to detect feature changes
  • Flag new features appearing and features lost
  • Schedule: First Monday of each month

Trigger Phrases

  • "How do I win featured snippets?"
  • "SERP feature audit for our keywords"
  • "Optimize our content for rich results"
  • "Which pages can win SERP features?"
  • "Steal featured snippets from competitors"
  • "PAA optimization plan"

平台分布

Codex

32.8%
按下载量换算24

Claude

32.02%
按下载量换算24

Cursor

20.17%
按下载量换算15

Gemini CLI

8.67%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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