internals

How MarkItDown Uses Magika for File Type Detection: Implementation Deep Dive

April 11, 2026 microsoft/markitdown ↗

MarkItDown creates a magika.Magika() instance at initialization and calls identify_stream() on every input file to generate ML-based content predictions that are merged with filename-based metadata, producing enriched StreamInfo objects that guide converter selection.

Microsoft’s MarkItDown leverages Magika, Google’s fast ML-based file-type detector, to enhance document conversion accuracy beyond simple extension checking. When you process a file through MarkItDown, the tool uses Magika to inspect raw bytes and confirm or correct the file type before selecting the appropriate converter

Have a question about this repo?

These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:

Share the following with your agent to get started:

curl -s "https://instagit.com/install.md"

Add to your MCP client configuration:

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}

Ask your agent:

"Use Instagit MCP to understand how microsoft/markitdown works."

Works with

Claude Codex Cursor VS Code OpenClaw Any MCP Client

Maintain an open-source project? Get it listed too →