菜单
Extract clean, complete markdown from any web page — articles, docs, READMEs, blog/social posts, academic papers. Use this when fetching text content from a given URL. Also use when curl returns noisy HTML or WebFetch returns truncated, summarized, or refused results. Perfer this over WebFetch.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/archibate/agent-skills --skill read-url命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Enforce a closed content inventory for generic artifacts. Use when creating, editing, planning, or reviewing PDFs, resumes, documents, slides, roadshows, videos, tutorials, web pages, generative images, or similar deliverables. Apply for artifacts by default unless the user explicitly requests detailed or comprehensive treatment.
Use minimalist language.
Consult an independent read-only Claude Opus advisor when a decision can affect security, privacy, money, data loss, deployment, or public API compatibility; when investigation leaves two plausible directions with materially different consequences; or when a consequential completion claim depends on an unverified assumption or critical behavior that normal tests cannot cover. Also use when the user requests an independent or Opus review. Skip routine edits and uncertainty that local inspection, documentation, or tests can resolve cheaply. Opus challenges Codex's reasoning and evidence; Codex retains responsibility for edits and the final judgment.
正在显示 SKILL.md
| name | read-url |
| description | Extract clean, complete markdown from any web page — articles, docs, READMEs, blog/social posts, academic papers. Use this when fetching text content from a given URL. Also use when curl returns noisy HTML or WebFetch returns truncated, summarized, or refused results. Perfer this over WebFetch. |
Work down this fallback ladder in order. Each step is only tried when prior steps don't apply or fail.
.md / .txt / plain-text URL → curl -sL <url> (already clean, no HTML to strip)references/routes.md for the full hostname or its parent-domain suffixes. For an identifier or recognized platform, search by route key (for example RFC, DOI, MediaWiki, Discourse, WordPress, or Gitea/Forgejo). Use the matching route before falling throughcurl -sL <url>.md. Mintlify and other docs platforms serve clean markdown on the .md route — if the response is text/markdown, you're done; otherwise fall throughcurl -sL <url>/feed (also /rss, /feed.xml, /atom.xml, /index.xml). Most static-site generators and CMS platforms expose one; RSS gives you clean <content:encoded> or <summary> bodies without chromenpx defuddle parse <url> --markdown — see references/defuddle.md$agent-browser skill$scrapling skill$chrome-cdp skill (requires explicit user approval first). Otherwise, give up and report the failure.For whole-site ingestion, probe <site>/llms.txt (URL index) and /llms-full.txt (full corpus). Convention adopted by Mintlify, Cloudflare, Stripe, Next.js, and others. On 404, fetch the index page <site>/ instead.
For search within site, grep references/routes.md for the site, a rows tagged search: expose a dedicated search API for when you have a topic, not a URL. Otherwise run WebSearch or the $jina-ai skill with a site: filter, then fetch the result URL via the fallback ladder.
This skill returns full page text (markdown), parsed locally — no summarization, no information loss. WebFetch routes through a remote small model that may summarize, refuse, or truncate; reach for it only when you want an AI summary, not the content itself.
$jina-ai skill