You know how you end up copy-pasting text from PDFs and it comes out mangled, or you need to extract content from a dozen different file types for RAG pipelines? This converts basically everything to markdown: documents, spreadsheets, presentations, images with OCR, audio transcriptions, URLs, even Jupyter notebooks and RSS feeds. Works as both a CLI and TypeScript library. The interesting part is it's pluggable, so you can write custom converters for weird formats, and it hooks into OpenAI, Anthropic, or any OpenAI-compatible API for the vision and transcription bits. Built by ara.so, who clearly got tired of dealing with this problem themselves.
npx skills add https://github.com/aradotso/trending-skills --skill markit-markdown-converter