---
source_url: "https://fetchmd.dev/?utm_source=openai"
title: FetchMD — Convert Any URL to Clean Markdown
mirrored_at: 2026-08-28T01:00:58.908Z
host: fetchmd.dev
cited_in_42a: true
mirror_canonical: "https://index.42a.ai/fetchmd.dev/index__q__utm_source_openai"
---

> **Original source:** https://fetchmd.dev/?utm_source=openai

API — Plans from $9/mo

Web pages, YouTube videos, PDFs, GitHub repos, research papers — all become clean Markdown in milliseconds. Cut token count by ~80%. Built for AI agents, RAG pipelines, and teams processing thousands of documents.

Web PagesYouTubePDFsGitHub ReposArxiv Papers

PASTE ANY URL. GET CLEAN MARKDOWN. ~80% FEWER TOKENS.

## HOW IT WORKS

01

### FETCH

We fetch the raw HTML from the target URL with proper headers and redirect handling.

02

### STRIP

Remove scripts, stylesheets, navigation, footers, ads, and other non-content elements.

03

### CONVERT

Transform HTML elements into clean Markdown syntax: headings, links, lists, code blocks.

04

### CLEAN

Normalize whitespace, decode entities, fix formatting, and deliver compact output.

## SUPPORTED SOURCES

One API. Every content type your agent needs.

### Web Pages

`POST /api/convert`

Any public URL → clean Markdown. Strips ads, nav, scripts, footers. Works with articles, docs, blogs, landing pages.

→

https://docs.anthropic.com/...

→

https://react.dev/learn

→

https://news.ycombinator.com

Counts as 1 conversion

### YouTube Videos

`POST /api/youtube`

Any YouTube video with captions → full transcript as Markdown. Timestamped or clean paragraphs. No API key needed.

→

youtube.com/watch?v=xxx

→

youtu.be/xxx

→

youtube.com/shorts/xxx

Counts as 1 conversion

### PDF Documents

`POST /api/pdf`

Link to any PDF → extracted text as Markdown. Research papers, reports, documentation, whitepapers. Up to 10MB.

→

arxiv.org/.../paper.pdf

→

example.com/report.pdf

→

Any direct PDF URL

Counts as 1 conversion

### GitHub Repositories

`POST /api/github`

GitHub repo URL → README + file tree as Markdown. Optionally include source files. Works with any public repo.

→

github.com/owner/repo

→

github.com/owner/repo/tree/main/src

→

github.com/owner/repo/blob/main/README.md

Counts as 1 conversion

### Arxiv Papers

`POST /api/convert`

Arxiv abstract pages → structured Markdown with title, authors, abstract, and links. Use /api/pdf for full paper text.

→

arxiv.org/abs/2301.07041

→

arxiv.org/pdf/2301.07041

Counts as 1 conversion

Bulk & Async

### Bulk Batches

`POST /api/bulk`

Process up to 500 URLs in a single call. Mix web pages, PDFs, GitHub links. Async jobs for fire-and-forget workflows.

→

200 docs pages at once

→

500 URLs from a sitemap

→

Async job + poll for results

Each URL = 1 conversion

## USE CASES

### AI AGENT BROWSING

Feed clean content to LLMs without wasting tokens on HTML tags, scripts, and styling. Get 80% more content in the same context window.

### RAG PIPELINES

Prepare web content for retrieval-augmented generation. Clean Markdown chunks better, embeds better, and retrieves better than raw HTML.

### DOCUMENTATION

Convert web pages to Markdown for documentation sites, wikis, or knowledge bases. Preserve structure without the HTML overhead.

### CONTENT MIGRATION

Move blog posts, articles, and pages from any platform to Markdown. Works with any public URL regardless of the CMS.

## PRICING

Scale from hundreds to thousands of conversions. [See all plans →](https://fetchmd.dev/pricing)

Free tier: 100 conversions/month — no credit card required

Scale Your Pipeline

## Ready to Go Beyond 100/Month?

Starter gets you 2,000 conversions. Pro gets you 10,000 with bulk batching and a priority queue. REST API, MCP server for Claude Desktop, and a 7-day free trial on every paid plan.