Token导航 LogoToken导航

Agent Skill · 研究检索

competitor-site-analysis(竞争对手网站分析)

competitor-site-analysis 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。

competitor-site-analysis 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,权限为操作浏览器;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 51,787。

总安装 574周安装 23GitHub Stars 35下载 186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:competitor-site-analysis(竞争对手网站分析)
来源仓库:https://github.com/superamped/ai-marketing-skills
仓库路径:skills/competitor-site-analysis
安装命令:
npx skills add https://github.com/superamped/ai-marketing-skills --skill competitor-site-analysis
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/superamped/ai-marketing-skills --skill competitor-site-analysis

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。

快速提取竞品网站五大核心模块信息生成标准化概况卡片。包含导航结构定价模型客户证言等关键元素完整性检查。如发现登录墙或异常状态立即终止流程保障数据安全合规性。仅需提供企业全称与主站URL即可完成初步信息抽取任务。competitor-site-analysis 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/competitor-site-analysis

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/superamped/ai-marketing-skills --skill competi…;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

    它需要操作浏览器,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

competitor-site-analysis 怎么安装?

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

competitor-site-analysis 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。

competitor-site-analysis 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

competitor-site-analysis 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。

competitor-site-analysis 用的人多吗?

公开数据是累计安装 574 次,近一周安装 23 次,下载 186 次,GitHub 35 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。

competitor-site-analysis 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
competitor-site-analysis

竞争对手网站分析 · 本页

57435操作浏览器未明确有提醒

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

Competitor Site Analysis

Usage

Use for a quick website data pull on a competitor, or as a first step before deeper analysis. Produces a structured 5-section profile from publicly available website information.

Process

Step 1: Gather Inputs

Ask the user for:

  1. Competitor name — the company to analyze
  2. Competitor URL — their homepage URL
  3. Specific sections to focus on (optional) — default: all 5

Step 2: Validate & Fetch

Confirm you have competitor name and URL. Fetch the homepage to verify it's reachable.

If the site is unreachable, behind a login wall, or returns errors — tell the user and stop.

Step 3: Fetch Website Pages

Fetch up to 6 pages. Adapt URLs based on what the site actually has — not every site uses the same paths.

PageTypical URLs to TryWhat to Extract
Homepage/Tagline, headline, value prop, target audience signals, trust badges, social proof
Pricing/pricing, /plansModel, tiers, price points, value metric, free tier, annual discount, trial, feature-by-tier gating
Features/features, /productKey features, integrations, platform capabilities
About/about, /about-us, /companyFounded, team size signals, stage/funding signals, mission
Customers/customers, /case-studiesCustomer logos, testimonials, case study count, industries served
Careers/careers, /jobs, /about/careersOpen roles by department, total headcount signals, strategic hiring patterns

If a page doesn't exist (404, redirect to homepage), skip it and note "not found" for that section. Don't guess.

Step 4: Structure Findings

Organize extracted data into 5 sections:

Section 1: Company Overview
  • Tagline (their words, not yours)
  • Founded / age signals
  • Stage / funding signals (bootstrapped, seed, Series X — based on about page, team size, office count)
  • Team size signals (exact if stated, otherwise estimate from about page)
  • Headquarters / markets
  • Primary GTM motion: PLG / sales-led / content-led / community-led (based on homepage CTA — "Start free" vs. "Book a demo" vs. both)
  • Partnerships / integrations as distribution (from features page or homepage)
Section 2: Product & Positioning
  • Homepage headline (exact quote)
  • Value proposition (what they promise in 1-2 sentences)
  • Target audience (who the site speaks to)
  • Key features (top 5-8 from features page)
  • Integrations (notable ones)
  • Product-led vs. sales-led signals (self-serve signup? demo request? both?)
  • Free trial / freemium availability
Section 3: Pricing & Feature Monetization
  • Pricing model (per-seat, usage-based, flat, freemium, custom)
  • Tier breakdown with price points
  • Value metric (what they charge per unit of)
  • Annual discount percentage
  • Free tier — what's included, what's limited
  • Enterprise tier — custom pricing or listed?

Feature monetization — if the pricing page shows feature-by-tier breakdowns:

  • Which features are free vs. gated?
  • What's the key unlock at each paid tier? (the feature that justifies the upgrade)
  • What do they gate behind enterprise? (reveals what they consider highest-value)
  • Any usage limits that force upgrades? (seats, API calls, storage, etc.)

If pricing is not public, note "not public — sales-led motion" and any signals about pricing level.

Section 4: Social Proof & Trust
  • Customer logos (notable names)
  • Testimonial highlights (1-2 strongest quotes)
  • Case study count and themes
  • Trust signals (G2 badges, security certs, compliance badges)
  • Community size (if visible — Slack, Discord, forum)
  • Social media follower counts (if visible on site)
Section 5: Hiring Signals
  • Total open roles (if listed)
  • Roles by department — engineering, sales, marketing, product, customer success, etc.
  • Notable hires or role types that signal strategic direction (e.g., "ML Engineer" = investing in AI, "Enterprise AE" = moving upmarket, "Developer Advocate" = building community)
  • Hiring velocity signals — "urgently hiring", multiple roles in one department

If no careers page exists, note "no public careers page" and move on.

Output Format

Present the 5-section structured website profile:

# Competitor Site Analysis: [Name]

**URL:** [url]
**Date:** [current date]

## 1. Company Overview
[structured findings]

## 2. Product & Positioning
[structured findings]

## 3. Pricing & Feature Monetization
[structured findings]

## 4. Social Proof & Trust
[structured findings]

## 5. Hiring Signals
[structured findings]

Rules

  • Never invent data — if a page is inaccessible or info isn't visible, mark "not found."
  • Never fetch more than 6 pages per competitor.
  • Never guess at pricing — if it's not public, say so.
  • Don't add strategic interpretation — this skill extracts, it doesn't analyze. Save strategic assessment for competitor-landscape.
  • Keep extraction factual. Quote headlines, list features, note prices. Don't editorialize.
  • If key pages (pricing, features) are missing or inaccessible, flag that the profile will have gaps.
  • If the site is heavily JS-rendered and content is sparse, suggest using Playwright MCP if available.

平台分布

Codex

35.85%
按下载量换算67

Claude

27.51%
按下载量换算51

Cursor

19.49%
按下载量换算36

Gemini CLI

9.37%
按下载量换算17

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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