Converts documents and files such as PDF, Word, PowerPoint, Excel, HTML, CSV, images, audio, ZIP and EPub into Markdown using markitdown.
- What it does
- This skill instructs an agent to run the markitdown command-line tool through uvx to convert supported file types into Markdown. It covers documents, web and data formats, media with EXIF/OCR or transcription, ZIP archives, YouTube URLs and EPub, and can write output to stdout or a file. Options include extension, MIME type and charset hints, Azure Document Intelligence for difficult PDFs, and third-party plugins. The resulting Markdown preserves headings, tables, lists and links for LLM processing or text analysis.
- When to use it
- Use it when you need to turn PDFs, Office files, HTML, CSV, JSON, XML, images, audio, ZIP archives, YouTube URLs or EPubs into Markdown text. It is suited to preparing documents for LLM processing or text analysis, and to extracting readable content from complex PDFs with Azure Document Intelligence.
- Requirements
- Requires the uvx runtime to run markitdown; no installation is needed as dependencies are cached on first run. Optional Azure Document Intelligence requires an endpoint and credentials. Network access is needed for uvx dependency retrieval and YouTube URLs. Ships no scripts; instructions only.
