by 安全扫描
OpenClaw
安全
high confidenceThe skill's declared purpose (website structure analysis) matches its instructions and resource use; it fetches and analyzes public site content and saves reports to the agent workspace, which is expected behavior.
评估建议
This skill appears coherent and does what it says: it will fetch public web pages, possibly interact with them (via the built-in browser), analyze structure/tech stack, and save Markdown reports and screenshots under /workspace/analysis. Before installing, consider: (1) only analyze sites you are allowed to crawl (respect robots.txt and site terms, and avoid private/internal sites with secrets), (2) the skill will store reports/screenshots in the agent workspace which may be accessible to other ...详细分析 ▾
✓ 用途与能力
The name/description describe website structure and competitive analysis and the SKILL.md contains detailed, purpose-aligned steps (collect pages, identify modules, infer tech stack, produce report). Required actions (fetch pages, screenshots, parse metadata) are consistent with the stated goal.
ℹ 指令范围
Instructions are focused on site crawling, DOM/metadata inspection, and report generation, which is appropriate. Minor note: the SKILL.md references a tool named 'extract_content_from_websites' (preferred) which is not declared in registry metadata — if that tool isn't available the skill falls back to the built-in 'browser' per instructions. The skill also directs saving reports/screenshots into /workspace/analysis, which is reasonable but means outputs and screenshots will be stored locally in the agent workspace.
✓ 安装机制
Instruction-only skill with no install spec or external downloads; nothing is written to disk by an installer and no third-party packages are pulled during install.
✓ 凭证需求
No environment variables, credentials, or external config paths are requested. The level of access (HTTP fetching and saving to /workspace) is proportionate to a web analysis tool.
✓ 持久化与权限
always is false and the skill does not request to be force-included. Autonomous invocation (default platform behavior) is allowed but not excessive in itself. The skill writes output only to its workspace path and does not attempt to modify other skills or global agent settings.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.02026/3/20
- Initial release of website-structure-analyzer. - Analyze and report website functional structure, including navigation, modules, page structure, tech stack, business model, and competitor differences. - Supports structured analysis for home, about, product/service, pricing, and footer navigation pages. - Saves detailed, structured reports and screenshots for reference. - Provides quick commands for competitor analysis, feature evaluation, tech stack audit, and page interaction review.
● 无害
安装命令 点击复制
官方npx clawhub@latest install website-structure-analyzer
镜像加速npx clawhub@latest install website-structure-analyzer --registry https://cn.clawhub-mirror.com
技能文档
分析任意网站的功能结构，输出完整的结构化分析报告。
工作流程
第一步：收集信息
使用以下工具按顺序收集网站数据：
工具优先级：
extract_content_from_websites— 首选，直接提取页面内容browser(OpenClaw 内置) — 用于截图和交互式操作
必抓页面：
- 主页 (/)
- 关于我们 (/关于 或 /关于-us)
- 产品/服务页 (/products 或 /services)
- 定价页 (/pricing)
- 底部导航（联系方式、招聘、法律条款等）
提取内容：
- 导航菜单结构（所有顶层 + 二级菜单项）
- 页面标题和副标题
- 表单类型（注册、登录、联系、订阅等）
- 交互元素（按钮、功能模块）
- 技术栈特征（Meta generator、WTF、Cookie 提示）
- 页脚 链接结构
第二步：识别功能模块
根据内容，分类识别以下模块：
| 模块类型 | 识别特征 |
|---|---|
| 导航系统 | 顶部菜单、侧边栏、面包屑 |
| 用户系统 | 登录、注册、用户中心 |
| 内容展示 | 轮播图、文章列表、产品卡片 |
| 交互表单 | 联系表、订阅、评论、搜索 |
| 交易系统 | 定价、购物车、结算、支付 |
| 社区功能 | 论坛、评论、评分、分享 |
| 数据功能 | 图表、仪表盘、导出 |
| 媒体功能 | 视频、图库、下载 |
第三步：输出结构化报告
# 网站功能结构分析报告
- 网站地址
- 分析时间
- 整体评价
1. 信息架构
（导航树形图）2. 核心功能模块
（每个模块的功能描述）3. 页面结构
（主要页面及内容概要）4. 技术栈推断
（前端框架、CMS、电商平台等）5. 商业模式识别
（B2B/B2C/C2C/SaaS/内容付费等）6. 竞品差异点
（独特功能或设计亮点）7. 可借鉴之处
（适合自身业务参考的功能）
第四步：保存报告
- 分析完成后，将报告保存至
/workspace/analysis/[域名]_[日期].md - 如有截图，保存至
/workspace/analysis/screenshots/
快捷命令
- 竞品分析：「分析 [URL] 的功能结构，作为竞品参考」
- 功能评估：「评估 [URL] 的 [具体功能] 是否值得借鉴」
- 技术尽调：「判断 [URL] 使用的技术栈和开发成本」
- 页面审计：「列出 [URL] 所有可交互的表单和功能点」
注意事项
- 先抓主页和导航，抓到完整菜单结构后再抓子页面
- 注意识别 Single Page Application (SPA)，内容可能是 JS 动态渲染
- 隐私政策页可了解数据收集和第三方服务
- Robots.txt 可了解站点结构和抓取偏好
- 相似域名（/blog、/help、/docs）与主站技术栈通常一致
数据来源：ClawHub ↗ · 中文优化：龙虾技能库
OpenClaw 技能定制 / 插件定制 / 私有工作流定制
免费技能或插件可能存在安全风险，如需更匹配、更安全的方案，建议联系付费定制