📦 Markdown Fetcher — 网页转Markdown

v1.0.0

一键抓取任意网页并输出干净、结构化的 Markdown 文本,方便后续编辑、存档或喂给 LLM,支持多解析引擎自动降级,无需写爬虫代码。

0· 180·0 当前·0 累计
jiaqi-guo-0114 头像by @jiaqi-guo-0114 (Jiaqi-Guo-0114)
下载技能包
最后更新
2026/3/17
0
安全扫描
VirusTotal
无害
查看报告
OpenClaw
安全
high confidence
This instruction-only skill is coherent with its stated purpose (convert webpages to Markdown) and only directs the agent to call public conversion endpoints or optionally install a Python scraper; nothing in the instructions requests unrelated credentials or privileged access.
评估建议
This skill is coherent but consider privacy and supply-chain cautions before use: any URL you convert will be fetched by the listed external services (markdown.new, defuddle.md, r.jina.ai or Scrapling), so avoid sending sensitive/private URLs or data. If you need the Scrapling fallback, inspect the Scrapling project (https://github.com/D4Vinci/Scrapling) before running 'pip install' and install it in a virtualenv or isolated environment. If you require strict confidentiality, prefer local tools ...
详细分析 ▾
用途与能力
The name and description match the runtime instructions: try hosted Markdown conversion endpoints (markdown.new, defuddle.md, r.jina.ai) and, if needed, fall back to a GitHub Python scraper (Scrapling). All required actions are proportionate to converting webpages to Markdown.
指令范围
SKILL.md confines the agent to fetching given URLs via specified services and (optionally) running 'pip install scrapling' and 'scrapling -u "URL"'. It does not instruct the agent to read local files, environment variables, or send data to unexpected endpoints beyond the listed services.
安装机制
There is no formal install spec, but the instructions recommend 'pip install scrapling' as a last-resort. Installing a Python package is a reasonable fallback, but pip installs execute code on the host and should be treated as a moderate-risk operation—verify the package source and run it in an isolated environment if needed.
凭证需求
The skill requests no environment variables, credentials, or config paths. This is appropriate for a web-to-Markdown converter that uses public endpoints or an optional local scraper.
持久化与权限
The skill is instruction-only, not always-enabled, and does not request persistent system presence or modify other skills. Autonomous invocation is allowed (platform default) but is not combined with other concerning privileges.
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

latestv1.0.02026/3/17

Initial release

无害

安装命令

点击复制
官方npx clawhub@latest install markdown-fetcher
镜像加速npx clawhub@latest install markdown-fetcher --registry https://cn.longxiaskill.com
数据来源ClawHub ↗ · 中文优化:龙虾技能库