markitdown

Convert diverse document formats to Markdown for AI processing.

Updated Mar 29, 2026

Search Tags:#pdf #markdown #ai #conversion #openai #openrouter

Ready to run

npx skills add https://github.com/trungkiet2005/ai_code_detection --skill markitdown-trungkiet2005

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: markitdown
Source: https://github.com/trungkiet2005/ai_code_detection/tree/main/.claude/skills/markitdown
Command: npx skills add https://github.com/trungkiet2005/ai_code_detection --skill markitdown-trungkiet2005

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

MarkItDown provides a robust pipeline to transform diverse documents into compact, AI-friendly Markdown, enabling efficient ingestion by LLMs and downstream analysis.

Core Features & Use Cases

AI-enhanced conversions: generate descriptions for images in PDFs, PPTX, and other formats using OpenRouter/OpenAI models.
Batch processing: parallel conversion of multiple files with optional plugin support for extensibility.
Format coverage: supports PDFs, Word, PowerPoint, Excel, HTML, CSV/JSON/XML, ZIP archives, images (with OCR), audio transcription, and YouTube transcripts.
Use Case: Convert a literature batch of PDFs to Markdown with optional index/catalog creation for literature reviews.

Quick Start

Install MarkItDown and run a simple conversion to verify it works.

Frequently Asked Questions about markitdown

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I convert PDF documents to Markdown for AI workflows?▼

Converting PDFs to Markdown for AI workflows is automated by extracting text and metadata while optionally using OpenAI or Azure Document Intelligence to process complex layouts and generate image descriptions. This produces AI-ready Markdown files efficiently for downstream LLM ingestion.

Can I batch convert multiple files to Markdown in parallel?▼

Batch converting multiple files to Markdown in parallel is supported, allowing simultaneous processing of mixed formats like Word, PowerPoint, and Excel. This enables efficient bulk transformations for research batches or data analysis pipelines, outputting structured Markdown files ready for LLM consumption.

Does converting documents to Markdown work with OpenRouter and OpenAI?▼

Markdown conversion works with both OpenRouter and OpenAI models to provide AI-enhanced processing, such as generating descriptions for images within PDFs and PPTX files. This integration enriches the extracted Markdown metadata, making the output more comprehensive for downstream analysis.

What file formats can be converted to Markdown for LLM ingestion?▼

Supported file formats for Markdown conversion include PDF, Word, PowerPoint, Excel, HTML, CSV, JSON, XML, ZIP archives, images with OCR, audio transcriptions, and YouTube transcripts. This broad format coverage ensures diverse document types can be transformed into compact, AI-friendly Markdown.

Do I need an OpenAI API key to extract metadata and generate descriptions during conversion?▼

An OpenAI or OpenRouter API key is required to enable AI-enhanced conversions like generating image descriptions and extracting complex metadata. Basic document-to-Markdown text extraction can proceed without it, but AI enhancements rely on these external providers to enrich the output.

What is the best way to process complex PDFs with images for literature reviews?▼

The best way to process complex PDFs with images for literature reviews is using optional Azure Document Intelligence alongside OpenAI models for accurate extraction and description generation. This creates structured Markdown with optional index or catalog creation, streamlining the review process.