Back to Details

markdown-fetch

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Markdown Fetch

Markdown Fetch

Efficiently fetch web content as clean Markdown using the markdown.new service.
使用markdown.new服务高效获取网页内容，输出为整洁的Markdown格式。

Why Use This

为什么使用本工具

  • 80% fewer tokens than raw HTML
  • 5x more content fits in context window
  • No external dependencies or parsing libraries needed
  • Three-tier conversion (Markdown-first, AI fallback, browser rendering)
  • 比原始HTML的token数量少80%
  • 上下文窗口可容纳的内容量提升5倍
  • 无需外部依赖或解析库
  • 三层转换机制（Markdown优先、AI兜底、浏览器渲染）

Triggering

触发条件

This skill should trigger automatically when:
  • User provides a URL (e.g., "Read https://example.com")
  • User asks to extract/fetch/analyze web content
  • User requests summarization of a webpage
  • User needs to process article/blog/documentation URLs
本skill会在以下场景自动触发：
  • 用户提供URL时（例如："Read https://example.com"）
  • 用户要求提取/获取/分析网页内容时
  • 用户请求对某个网页进行总结时
  • 用户需要处理文章/博客/文档类URL时

Quick Start

快速开始

bash
undefined
bash
undefined

Fetch any URL

Fetch any URL

scripts/fetch.sh "https://example.com"
scripts/fetch.sh "https://example.com"

Use browser rendering for JS-heavy sites

Use browser rendering for JS-heavy sites

scripts/fetch.sh "https://example.com" --method browser
scripts/fetch.sh "https://example.com" --method browser

Retain images in output

Retain images in output

scripts/fetch.sh "https://example.com" --retain-images
undefined
scripts/fetch.sh "https://example.com" --retain-images
undefined

Typical Usage Patterns

典型使用场景

When a user says:
  • "Read this article: https://..." → Use this skill to fetch the content
  • "Summarize https://..." → Fetch with this skill first, then summarize
  • "What does this page say: https://..." → Fetch the content
  • "Extract the text from https://..." → Use this skill
当用户提出以下需求时：
  • "阅读这篇文章：https://..." → 使用本skill获取内容
  • "总结https://..." → 先用本skill抓取内容，再进行总结
  • "这个页面讲了什么：https://..." → 抓取内容
  • "提取https://...的文本内容" → 使用本skill

Conversion Methods

转换方法

auto (default) - Try Markdown-first, fall back to AI or browser as needed
ai - Use Cloudflare Workers AI for conversion
browser - Full browser rendering for JS-heavy content
auto（默认） - 优先尝试Markdown转换，必要时兜底使用AI或浏览器渲染
ai - 使用Cloudflare Workers AI进行转换
browser - 针对重JS内容启用完整浏览器渲染

Options

参数选项

--method <auto|ai|browser>
- Conversion method
--retain-images
- Keep image references in output
--output <file>
- Save to file instead of stdout
--method <auto|ai|browser>
- 转换方法
--retain-images
- 在输出中保留图片引用
--output <file>
- 保存到文件而非输出到stdout

Output

输出

Returns clean Markdown with metadata:
markdown
---
title: Page Title
url: https://example.com
method: auto
duration_ms: 725
fetched_at: 2026-03-07T12:00:00Z
---
返回带元数据的整洁Markdown内容：
markdown
---
title: Page Title
url: https://example.com
method: auto
duration_ms: 725
fetched_at: 2026-03-07T12:00:00Z
---

Content here...

Content here...

undefined
undefined

When to Use

适用场景

  • Extracting articles, documentation, or blog posts
  • Building RAG pipelines with web content
  • Summarizing web pages
  • Fetching content for analysis
  • Converting sites to Markdown format
  • 提取文章、文档或博客帖子内容
  • 基于网页内容搭建RAG pipeline
  • 网页内容总结
  • 抓取内容用于分析
  • 将站点转换为Markdown格式

Implementation Notes

实现说明

The service handles:
  • Content negotiation (Accept: text/markdown)
  • Cloudflare Workers AI conversion
  • Browser rendering for dynamic content
  • Automatic fallback between methods
本服务支持：
  • 内容协商（Accept: text/markdown）
  • Cloudflare Workers AI转换
  • 动态内容的浏览器渲染
  • 多方法自动兜底切换