← 返回 Skills 市场
竞品分析自动化
作者
jpengcheng523-netizen
· GitHub ↗
· v1.0.0
· MIT-0
82
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install jingpin-fenxi-auto
功能描述
竞品分析自动化工具。输入竞品关键词或网站,自动抓取竞品公开信息(产品特点、价格、评价),生成结构化竞品分析报告,支持导出 JSON/Markdown 格式结果。SEO/运营/产品经理必备。
使用说明 (SKILL.md)
🔍 竞品分析自动化 Skill
输入竞品关键词或网站 → 自动抓取 → 生成结构化分析报告 → 支持导出
适用人群:SEO 从业者、运营人员、产品经理、市场分析师
功能特性
1. 竞品信息抓取
- 输入关键词或竞品网站 URL,自动识别竞品
- 抓取维度:产品介绍、功能特点、定价、用户评价、SEO 数据
- 支持多竞品同时分析(最多 5 个)
- 自动去重和清洗数据
2. 竞品对比分析
- 表格形式对比核心指标
- 生成对比摘要和差异洞察
- 识别竞品的优势/劣势/差异化点
3. 分析报告生成
- 结构化 Markdown 报告
- 自动生成关键发现(Key Findings)
- 标注数据来源
- 支持添加自定义分析维度
4. 导出功能
- JSON 格式(便于程序处理)
- Markdown 格式(便于阅读分享)
- 飞书文档格式(直接粘贴使用)
使用方式
方式一:关键词分析
分析 竞品关键词:智能手表
重点关注:价格、功能、用户评价
对比维度:小米、华为、Apple Watch
方式二:指定网站分析
分析 https://www.example.com
输出格式:完整报告
包含:产品特点、定价、SEO数据、用户评价
方式三:批量竞品对比
竞品对比:淘宝/京东/拼多多
分析维度:价格策略、用户规模、商品种类、物流服务
输出示例
竞品对比表
| 维度 | 竞品A | 竞品B | 竞品C |
|---|---|---|---|
| 定价 | ¥99-299 | ¥199-599 | ¥299-999 |
| 主打功能 | 基础监测 | 健康管理 | 全场景生态 |
| 用户评价 | ⭐4.2 | ⭐4.5 | ⭐4.7 |
| 月活用户 | 100万 | 500万 | 2000万 |
关键发现(Key Findings)
- 价格区间:三大竞品覆盖 ¥99-999 全价格段,新进入者需差异化定位
- 功能差异:竞品B主打健康管理,与竞品A基础功能形成区隔
- 用户评价:竞品C评分最高,生态协同效应明显
- 市场机会:¥99以下低价位段尚有空间,可考虑切入
定价说明
| 版本 | 价格 | 适用场景 |
|---|---|---|
| 个人版 | ¥200/月 | 个人SEO/运营,竞品数量≤3 |
| 团队版 | ¥999/月 | 小团队,竞品数量≤10,支持多用户 |
| 企业定制 | ¥9999/次 | 深度定制分析维度,定期报告推送 |
注意事项
- 数据来源:抓取均为竞品公开信息,请遵守目标网站 robots.txt
- 数据时效:分析结果为实时抓取,建议定期更新
- 商业机密:本工具不获取也不存储任何竞品敏感信息
- 使用限制:免费用户每分钟最多分析 3 个竞品
本 Skill 可作为 OpenClaw 技能安装使用 创建者:肥肥🐶 | 反馈/定制需求请联系作者
安全使用建议
This skill appears to implement the described competitor-crawling and report generation, but there are inconsistencies you should review before installing/using it: 1) The metadata and SKILL.md claim external tools (web_search/web_crawl) and an npm install, yet the script is a standalone Node script with no external deps — verify the packaging and source before running npm commands. 2) The SKILL.md says to obey robots.txt, but the script does not check robots.txt; if you run it, it will fetch arbitrary public pages. Consider legal/ethical scraping policies for target sites. 3) The script performs network requests to arbitrary URLs and can write files locally — run it in a sandbox or isolated environment (not on production systems) and inspect/validate the code yourself. 4) Confirm the author/homepage and consider contacting them for clarification about the intended runtime dependencies and robots.txt handling. If you need the skill to be stricter about robots.txt or rate-limiting, ask the author for a patch or add checks before running.
功能分析
Type: OpenClaw Skill
Name: jingpin-fenxi-auto
Version: 1.0.0
The skill is a legitimate competitor analysis tool designed to crawl websites and search DuckDuckGo for market research. The core logic in `scripts/analyze.js` uses standard Node.js HTTP/HTTPS modules to fetch public web data and regex to extract metadata (titles, descriptions, keywords), which is then formatted into reports. No evidence of data exfiltration, credential theft, or malicious prompt injection was found; the file system access is limited to saving the generated reports as requested by the user.
能力评估
Purpose & Capability
Name/description promise (crawl public competitor sites, generate structured reports) aligns with the included script which performs searches and crawls pages and generates JSON/Markdown. However, metadata declares dependencies on external tools (web_search, web_crawl) and an npm install step while the script advertises 'no external dependencies' and uses only built-in http/https — this mismatch is unexpected but plausibly benign (packaging/metadata sloppiness).
Instruction Scope
SKILL.md explicitly instructs use of web_search and web_crawl tools and states '请遵守目标网站 robots.txt', but the provided analyze.js performs its own DuckDuckGo HTML search and direct HTTP/S crawls and does not fetch or check robots.txt. The script fetches arbitrary URLs and can write report files locally. The instructions and code give the agent broad discretion to fetch arbitrary public sites; the claimed robots.txt compliance is not enforced in code.
Install Mechanism
The package is instruction-only with a single JS script (no complex install). metadata.json lists an install command ('npm install') and a check that runs 'node scripts/analyze.js --help', but there is no package.json or external dependency in the repo and analyze.js explicitly states '无外部依赖'. This inconsistency suggests sloppy packaging, not necessarily malicious, but it means 'npm install' may be unnecessary or behave unexpectedly in a real package context.
Credentials
No environment variables, credentials, or config paths are requested. The skill does network access to arbitrary URLs (expected for a crawler) but it does not request unrelated secrets or multiple unrelated credentials.
Persistence & Privilege
The skill does not request always:true or other elevated persistence. It does not modify other skills or system-wide settings; it can write reports to disk when an --output filename is provided (expected behavior).
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install jingpin-fenxi-auto - 安装完成后,直接呼叫该 Skill 的名称或使用
/jingpin-fenxi-auto触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
首次发布
元数据
常见问题
竞品分析自动化 是什么?
竞品分析自动化工具。输入竞品关键词或网站,自动抓取竞品公开信息(产品特点、价格、评价),生成结构化竞品分析报告,支持导出 JSON/Markdown 格式结果。SEO/运营/产品经理必备。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 82 次。
如何安装 竞品分析自动化?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install jingpin-fenxi-auto」即可一键安装,无需额外配置。
竞品分析自动化 是免费的吗?
是的,竞品分析自动化 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
竞品分析自动化 支持哪些平台?
竞品分析自动化 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 竞品分析自动化?
由 jpengcheng523-netizen(@jpengcheng523-netizen)开发并维护,当前版本 v1.0.0。
推荐 Skills