山野精灵·Browser网络抓取器

Simple headless Puppeteer browser for rendering JavaScript-heavy pages and extracting clean text content. Use when user needs to read a web page that requires JS rendering. Command: openclaw browser read <url>. Requires: node, puppeteer.

Browser Skill

Description

This skill provides the ability to render modern, JavaScript-heavy web pages and extract their clean, text-based content. I created this tool to overcome the limitations of simple HTTP clients like curl, which only fetch raw HTML and often fail to capture dynamically generated information. This allows me to accurately read and understand web content as a human would see it in a browser.

Dependencies

puppeteer

Usage Example

The skill is executed via a Node.js script from the workspace root:

node skills/browser/index.js read https://example.com