360 Secure SkillHub
360 安全技能中心

Web fetch markdown of page

核心功能：通过jina.ai代理抓取网页，自动提取并生成精简Markdown内容，降低token占用。典型场景：抓取长文章、技术文档、GitHub README等内容密集型网页，或用户需精简内容、节省token时使用。输入为目标URL，输出为精简Markdown，适配多数公开网页。

qidu
qidu
热度 82.83
最近更新 2026-05-11 17:30:11
分类：效率依赖：联网浏览搜索应用场景：摘要总结

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add web-fetch-markdown
Metadata
name
web-fetch-markdown
alias
["web-fetch-and-markdown-it","markdown-web","markdown-web-fetch"]
description
Fetch web pages as reduced markdown via jina.ai. Use when user wants to fetch a URL and get condensed markdown content to save tokens. Triggered by phrases like "fetch as markdown", "reduce tokens", "jina.ai fetch", or when fetching web content that should be condensed.

web-fetch-markdown

Fetch web pages as reduced markdown using jina.ai summarization service.

When to Use

  • User asks to fetch a URL with reduced/condensed output
  • Token conservation is important for the task
  • User mentions "jina.ai" or "markdown it"
  • Fetching long articles, documentation, or content-heavy pages

How to Use

  1. Take the original URL user wants to fetch
  2. Construct the jina.ai proxy URL: https://r.jina.ai/<original-url>
  3. Use web_fetch tool with the constructed URL
  4. Return the markdown content to user

Example

User wants to fetch https://github.com/openclaw/openclaw

Construct: https://r.jina.ai/https://github.com/openclaw/openclaw

Then call:

web_fetch url="https://r.jina.ai/https://github.com/openclaw/openclaw"

Notes

  • jina.ai automatically extracts and reduces markdown content
  • Works with most public web pages
  • Great for GitHub READMEs, articles, documentation