← 返回 Skills 市场
Full Webpage Screenshot
作者
lifelooploop
· GitHub ↗
· v1.0.2
· MIT-0
157
总下载
1
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install full-webpage-screenshot
功能描述
Capture full-page screenshots of websites with lazy-load support. Use when: user wants to screenshot a webpage, take a website screenshot, capture a full pag...
使用说明 (SKILL.md)
Full Webpage Screenshot
Capture complete webpage screenshots using Puppeteer, including content that loads dynamically on scroll.
When to Use
✅ USE this skill when:
- "Screenshot this website"
- "Take a screenshot of example.com"
- "Capture the full page"
- "Get a visual of this webpage"
- "截个网页图"
- "给这个网站截屏"
When NOT to Use
❌ DON'T use this skill when:
- Screenshotting local files → use system screenshot tools
- Capturing specific elements only → use browser DevTools
- Video/screenshots of interactions → use screen recording
- Need for authenticated pages → manual browser required
Setup
First-time setup (installs Puppeteer):
cd ~/.openclaw/skills/full-webpage-screenshot/scripts
npm install
Commands
Basic Screenshot
cd ~/.openclaw/skills/full-webpage-screenshot/scripts
node screenshot.js "https://example.com" ~/workspace/screenshot.png
Custom Viewport (Mobile/Desktop)
# Mobile view
cd ~/.openclaw/skills/full-webpage-screenshot/scripts
VIEWPORT_WIDTH=375 VIEWPORT_HEIGHT=812 node screenshot.js "https://example.com" mobile.png
# Desktop HD
VIEWPORT_WIDTH=1920 VIEWPORT_HEIGHT=1080 node screenshot.js "https://example.com" desktop.png
Slow-Loading Pages
cd ~/.openclaw/skills/full-webpage-screenshot/scripts
WAIT_AFTER=5000 node screenshot.js "https://slow-site.com" screenshot.png
Options
| Variable | Default | Description |
|---|---|---|
VIEWPORT_WIDTH |
1280 | Browser viewport width |
VIEWPORT_HEIGHT |
800 | Browser viewport height |
SCROLL_DELAY |
100 | Delay between scroll steps (ms) |
WAIT_AFTER |
2000 | Wait after page load (ms) |
Output
Returns JSON:
{
"success": true,
"path": "/path/to/screenshot.png",
"width": 1280,
"height": 4352
}
Quick Responses
"Screenshot this site"
cd ~/.openclaw/skills/full-webpage-screenshot/scripts
node screenshot.js "$URL" ~/workspace/screenshot.png
"Mobile screenshot"
cd ~/.openclaw/skills/full-webpage-screenshot/scripts
VIEWPORT_WIDTH=375 VIEWPORT_HEIGHT=812 node screenshot.js "$URL" mobile.png
Notes
- Requires Node.js 18+
- First run needs
npm installin scripts/ - Automatically scrolls to trigger lazy-loaded content
- Supports all modern web features (JS, CSS, fonts)
安全使用建议
This skill appears to be what it claims: a Puppeteer script that captures full-page screenshots. Before installing, note that npm install will download Puppeteer and a Chromium binary (large download and native code) into the skill directory — consider running it in an isolated environment or container if you prefer. The script will open arbitrary URLs and write the screenshot file to whatever output path you provide, so avoid pointing it at authenticated internal sites or sensitive pages without appropriate safeguards. If you're cautious, review package-lock.json (it uses registry.npmjs.org) and run the install in a sandboxed environment (or use npm ci from a vetted lockfile).
能力评估
Purpose & Capability
Name/description match the code and instructions: the skill uses Node + Puppeteer to capture full-page screenshots and scroll to trigger lazy-loaded content. Required binary (node) and the bundled scripts/package.json (puppeteer dependency) are appropriate.
Instruction Scope
SKILL.md's runtime instructions are limited to installing Node modules in the skill directory and running screenshot.js with a target URL and output path. It does not instruct reading unrelated system files, accessing secrets, or sending data to external endpoints beyond loading the target web page.
Install Mechanism
There is no formal install spec, but SKILL.md instructs running npm install in the skill's scripts directory. This will pull puppeteer from the public npm registry and (as is typical) download a Chromium build during install — a moderate-risk action that is expected for Puppeteer-based tools. All package references in package-lock.json point to npmjs.org (no suspicious hosts).
Credentials
No credentials, config paths, or sensitive environment variables are required. Optional env vars (VIEWPORT_WIDTH/HEIGHT, SCROLL_DELAY, WAIT_AFTER) are appropriate for controlling screenshot behavior.
Persistence & Privilege
The skill is not forced-always, is user-invocable, and does not request persistent system-wide privileges or modify other skills' configurations. It runs as a normal CLI script when invoked.
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 full-webpage-screenshot - 安装完成后,直接呼叫该 Skill 的名称或使用
/full-webpage-screenshot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
v1.0.2
修复所有路径引用为 full-webpage-screenshot
v1.0.1
更新技能名称和路径引用
v1.0.0
支持懒加载的完整网页截图工具,自动滚动触发动态内容
元数据
常见问题
Full Webpage Screenshot 是什么?
Capture full-page screenshots of websites with lazy-load support. Use when: user wants to screenshot a webpage, take a website screenshot, capture a full pag... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 157 次。
如何安装 Full Webpage Screenshot?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install full-webpage-screenshot」即可一键安装,无需额外配置。
Full Webpage Screenshot 是免费的吗?
是的,Full Webpage Screenshot 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Full Webpage Screenshot 支持哪些平台?
Full Webpage Screenshot 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 Full Webpage Screenshot?
由 lifelooploop(@lifelooploop)开发并维护,当前版本 v1.0.2。
推荐 Skills