Yapit – PDF and webpage reader with TTS that doesn't suck
Vision-LLM preprocessing fixes math and layout garbling that breaks Speechify.

Markdownload exists, but direct File System Access API write avoids cloud sync.
Obsidian users, researchers, developers saving documentation
Markdownload · Obsidian Web Clipper · Readwise Reader
Vision-LLM preprocessing fixes math and layout garbling that breaks Speechify.
Nice, focused product: site-specific extraction rules (CSS selectors/metadata overrides), edge-first delivery (<500ms p99) and SDKs for Node/Python make it quick to drop into an LLM pipeline and claim 40–60% token savings. That said, HTML→Markdown is a crowded niche (Pandoc, Jina, Firecrawl and dozens of scrapers already exist), so Klovr needs clearer differentiation — e.g. demonstrable extraction accuracy, enterprise-grade rule sharing, or unique model-aware trimming — to move beyond 'handy utility'.
PDF-to-Markdown for LLMs when JinaAI and Firecrawl already exist.
Markdown viewer with Mermaid and LaTeX when browsers already render MD.
CPU-only VLM OCR beats Tesseract on layout without needing CUDA or cloud APIs.
Splits LLM Markdown into chat-sized WhatsApp messages while preserving lists, links, emails, tables and even Spanish punctuation. It applies a priority chain of processors — structural splits first, semantic fallbacks — and ships with zero dependencies plus 100% test coverage, which makes it a pragmatic, focused tool for messaging pipelines.