AgentMD

Documents in. Clean markdown out.

An agent-native conversion API. Send a PDF, DOCX, HTML file, or any URL — get back LLM-ready markdown with tables preserved and boilerplate stripped. Use it as a REST API or plug it straight into your agent as an MCP server. Priced per call, no seats.

50 free conversions, no card. Agents can self-provision: POST /api/v1/keys/free with {"email": "..."}. After that it's $0.002 per successful conversion, billed monthly.

REST: convert a URL

curl -X POST https://www.getagentmd.com/api/v1/convert \
  -H "Authorization: Bearer <your-key>" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://en.wikipedia.org/wiki/Markdown"}'

REST: upload a file

curl -X POST https://www.getagentmd.com/api/v1/convert \
  -H "Authorization: Bearer <your-key>" \
  -F "file=@report.pdf"

Or send JSON with base64 + filename. Responses include markdown, meta (format, pages, token estimate), and warnings.

MCP: give it to your agent

One-line install, no hosting required:

claude mcp add agentmd \
  --env AGENTMD_API_KEY=<your-key> -- npx -y agentmd-mcp

Or point any MCP client at the hosted server over streamable HTTP:

{
  "mcpServers": {
    "agentmd": {
      "url": "https://www.getagentmd.com/api/mcp",
      "headers": { "Authorization": "Bearer <your-key>" }
    }
  }
}

Exposes convert_url_to_markdown, convert_file_to_markdown, and convert_document_to_markdown.

Supported today

  • PDF (text extraction; OCR for scanned PDFs is on the roadmap)
  • DOCX (headings, lists, and tables preserved)
  • HTML & URLs (main-article extraction via Readability)
  • Markdown & plain text (normalized passthrough)
  • Up to 25 MB per document

For machines