Docs

API

Prepend this host to a URL to convert it. Batch, extract, jobs, and deep crawl need an API key.

Quick Start

Get raw Markdown

curl -H "Accept: text/markdown" https://mdget.dev/https://example.com

Get JSON output

curl "https://mdget.dev/https://example.com?raw=true&format=json"

Batch conversion

curl -X POST https://mdget.dev/api/batch \
  -H "Authorization: Bearer API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"urls":["https://example.com"]}'

Structured extraction

curl -X POST https://mdget.dev/api/extract \
  -H "Authorization: Bearer API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"strategy":"css","url":"https://example.com","schema":{"fields":[{"name":"title","selector":"h1","type":"text","required":true}]}}'

Deep crawl

curl -X POST https://mdget.dev/api/deepcrawl \
  -H "Authorization: Bearer API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"seed":"https://example.com/docs","stream":true}'

API Reference

Routes
`GET /{url}`	Convert a single URL to Markdown
`GET /api/stream`	SSE progress for single-URL conversion (step / done / fail)
`POST /api/batch`	Convert up to 10 URLs (requires API_TOKEN)
`POST /api/extract`	Structured extraction (css / xpath / regex)
`POST /api/jobs`	Create, query, stream, and run jobs
`POST /api/deepcrawl`	BFS / BestFirst deep crawl with filters and scoring
`GET /api/health`	Health check and operational metrics
`GET /api/og`	Share-image generator
`GET /llms.txt`	AI-readable API description

Query Parameters

Parameter
`?raw=true`	Return raw Markdown (no HTML wrapper)
`?format=`	Output format (markdown \| html \| text \| json)
`?selector=.css`	Extract only matching CSS selector
`?force_browser=true`	Force headless browser rendering
`?engine=jina\|firecrawl\|cf`	Convert via specific engine (jina / firecrawl / cf)
`?no_cache=true`	Bypass cache, fetch fresh content
`?token=`	Public API token

Authentication

Response Headers

Header
`X-Markdown-Method`	native \| readability+turndown \| browser+readability+turndown \| jina
`X-Cache-Status`	HIT \| MISS
`X-Source-URL`	The original target URL