Stash — Smart Capture
We consume more information than ever—articles, threads, and videos—yet most of it gets lost in bookmark graveyards or forgotten browser tabs. I built Stash to break this cycle. The name comes from the git stash command: just as developers temporarily stash their code to switch contexts without losing progress, this tool lets you instantly save valuable content you stumble upon without breaking your current workflow. It allows you to safely set information aside and seamlessly return to it when you are ready, transitioning from mindless scrolling to intentional saving.
More than just a simple web clipper, Stash is a “knowledge collage tool” designed to weave these scattered fragments together. Available as a Chrome Extension and web app, it lets you capture content from any platform—like Instagram, LinkedIn, YouTube, or Twitter—with a single click. AI automatically enriches, tags, and categorizes everything you save, which is then ingested into a RAG (Retrieval-Augmented Generation) pipeline. You are not just storing static links; you are building an intelligent personal database where you can easily pull exact insights whenever you need them, transforming scattered internet noise into a connected knowledge base you truly own.
Tech Stack
Features
How users retrieve insights from their saved captures. A hybrid RAG system that automatically switches between Long Context (full library injection) and semantic embedding search based on library size—optimizing for both accuracy and cost.
How I prevent AI Chat from becoming a free ChatGPT. 4 server-enforced layers—rate limiting, quota, per-request caps, and scope enforcement—to keep LLM token costs bounded and conversations within the user's collage.
Why common class names are a silent attack surface for injected UI. A content-script modal that rendered fine everywhere except Brave—and the CSS cascade-origin rule that explains why Shadow DOM isn't the fortress you think it is.