How do I install the Markdown New skill?

Run npx skillmds@latest add cyanheads/markdown-new in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Markdown New safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Markdown New?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Markdown New free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Markdown New?

cyanheads (@cyanheads) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Markdown New by cyanheads — Agent Skill

Name: Markdown New
Author: cyanheads

Markdown New

Fetch the full, unabridged content of a URL as clean Markdown. Use when you need the actual content of a web page — not a summary or extraction, but the real text, tables, and structure. Covers markdown.new, Cloudflare Workers AI toMarkdown, Markdown for Agents content negotiation, and Browser Rendering.

cyanheads Updated Sep 17, 2026

File contents

markdown.new

Free URL-to-Markdown conversion. No auth required. Converts any public URL to clean Markdown.

Quick Usage

curl -s 'https://markdown.new/' \
  -H 'Content-Type: application/json' \
  -d '{"url": "https://example.com"}' | jq -r '.content'

That's it. For most cases, this is all you need.

When to Use

You need the actual content of a web page, not a lossy summary
You want the full text, tables, code blocks, and structure preserved
The page has content that could be missed by partial extraction
You need to read documentation, articles, or data-heavy pages faithfully

API Details

Parameters

Parameter	Values	Default	Notes
`url`	Any public URL	—	Required
`method`	`auto`, `ai`, `browser`	`auto`	`auto` tries fast conversion first, falls back to headless browser. Use `browser` explicitly for JS-heavy SPAs.
`retain_images`	`true`, `false`	`false`	Keep image references in output

Request Formats

POST JSON (preferred):

curl -s 'https://markdown.new/' \
  -H 'Content-Type: application/json' \
  -d '{"url": "https://example.com"}'

URL prepend (browser/quick access):

https://markdown.new/https://example.com

Query parameters:

https://markdown.new/https://example.com?method=browser&retain_images=true

Response

{
  "success": true,
  "url": "https://example.com",
  "title": "Page Title",
  "content": "# Page Title\n\nMarkdown content...",
  "method": "Cloudflare Workers AI",
  "duration_ms": 306,
  "tokens": 939
}

The content field is what you want. Use jq -r '.content' to extract it directly.

Rate Limits

500 requests/day per IP
HTTP 429 when exceeded
Track via x-rate-limit-remaining response header

Edge Cases

Authenticated/paywalled pages: Only public URLs work.
JS-rendered SPAs: Use method=browser explicitly — auto may return incomplete content from faster tiers.
Very large pages: May be truncated. Origin responses over 2 MB may fail.
Rate limiting: 500/day is generous for ad-hoc use. If a task requires bulk fetching, space requests or switch to Cloudflare Workers AI toMarkdown directly (requires a Cloudflare account).
Images: Excluded by default. Set retain_images=true to keep image references.

For Building Applications

When the user is integrating URL-to-Markdown into their own code (not ad-hoc fetching):

Content Negotiation (Cloudflare Zones)

Sites on Cloudflare with "Markdown for Agents" enabled respond to Accept: text/markdown directly:

curl -s 'https://example.com' -H 'Accept: text/markdown'

Workers AI `toMarkdown`

For programmatic use at scale — converts HTML, PDF, and images to Markdown. Available as a Worker binding or REST API. Requires a Cloudflare account.

Binding:

const result = await env.AI.toMarkdown(
  [{ name: "page.html", blob: htmlBlob }],
  { conversionOptions: { html: { cssSelector: "main" } } }
);

REST:

curl https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/tomarkdown \
  -X POST \
  -H 'Authorization: Bearer {API_TOKEN}' \
  -F "files=@page.html"

---name: markdown-newdescription: Fetch the full, unabridged content of a URL as clean Markdown. Use when you need the actual content of a web page — not a summary or extraction, but the real text, tables, and structure. Covers markdown.new, Cloudflare Workers AI toMarkdown, Markdown for Agents content negotiation, and Browser Rendering.---# markdown.newFree URL-to-Markdown conversion. No auth required. Converts any public URL to clean Markdown.## Quick Usage```bashcurl -s 'https://markdown.new/' \ -H 'Content-Type: application/json' \ -d '{"url": "https://example.com"}' | jq -r '.content'```That's it. For most cases, this is all you need.## When to Use- You need the actual content of a web page, not a lossy summary- You want the full text, tables, code blocks, and structure preserved- The page has content that could be missed by partial extraction- You need to read documentation, articles, or data-heavy pages faithfully## API Details### Parameters| Parameter | Values | Default | Notes ||-----------|--------|---------|-------|| `url` | Any public URL | — | Required || `method` | `auto`, `ai`, `browser` | `auto` | `auto` tries fast conversion first, falls back to headless browser. Use `browser` explicitly for JS-heavy SPAs. || `retain_images` | `true`, `false` | `false` | Keep image references in output |### Request Formats**POST JSON** (preferred):```bashcurl -s 'https://markdown.new/' \ -H 'Content-Type: application/json' \ -d '{"url": "https://example.com"}'```**URL prepend** (browser/quick access):```https://markdown.new/https://example.com```**Query parameters:**```https://markdown.new/https://example.com?method=browser&retain_images=true```### Response```jsonc{ "success": true, "url": "https://example.com", "title": "Page Title", "content": "# Page Title\n\nMarkdown content...", "method": "Cloudflare Workers AI", "duration_ms": 306, "tokens": 939}```The `content` field is what you want. Use `jq -r '.content'` to extract it directly.### Rate Limits- **500 requests/day per IP**- HTTP 429 when exceeded- Track via `x-rate-limit-remaining` response header## Edge Cases- **Authenticated/paywalled pages**: Only public URLs work.- **JS-rendered SPAs**: Use `method=browser` explicitly — `auto` may return incomplete content from faster tiers.- **Very large pages**: May be truncated. Origin responses over 2 MB may fail.- **Rate limiting**: 500/day is generous for ad-hoc use. If a task requires bulk fetching, space requests or switch to Cloudflare Workers AI `toMarkdown` directly (requires a Cloudflare account).- **Images**: Excluded by default. Set `retain_images=true` to keep image references.## For Building ApplicationsWhen the user is integrating URL-to-Markdown into their own code (not ad-hoc fetching):### Content Negotiation (Cloudflare Zones)Sites on Cloudflare with "Markdown for Agents" enabled respond to `Accept: text/markdown` directly:```bashcurl -s 'https://example.com' -H 'Accept: text/markdown'```### Workers AI `toMarkdown`For programmatic use at scale — converts HTML, PDF, and images to Markdown. Available as a Worker binding or REST API. Requires a Cloudflare account.**Binding:**```tsconst result = await env.AI.toMarkdown( [{ name: "page.html", blob: htmlBlob }], { conversionOptions: { html: { cssSelector: "main" } } });```**REST:**```bashcurl https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/tomarkdown \ -X POST \ -H 'Authorization: Bearer {API_TOKEN}' \ -F "files=@page.html"```

Frequently asked questions