mdream Stash

Stash webpages, PDFs, and Office files as clean Markdown. Organize, group, and export.

Save any article or page as clean Markdown, directly from Chrome. mdream Stash lives in the side panel. Open it, go to a normal website, and tap "Stash This Page" to capture the active tab as readable Markdown. Build a personal library: sort entries into groups, search by title or URL, and open any item to read, copy, or download the Markdown. BACKUP AND PORTABILITY Export your library as JSON (full backup including groups and metadata), as one combined Markdown file, or as llms.txt for AI-style workflows. Import the same formats: JSON supports replace (full restore) or merge; Markdown and llms.txt imports add pages to your current group (or Inbox when "All" is selected). PRIVACY AND DATA Your stash never leaves your device. Entries are stored only in Chrome's local extension storage. The extension does not send page content, URLs, or titles to any server operated by the developer. WHY PERMISSIONS The extension needs access to http and https pages so it can read the page HTML when you explicitly click "Stash This Page." Chrome does not treat a side-panel button as an "active tab" gesture, so host access is required for that one action—not for background scanning. Scripting runs a one-time read of the DOM; storage saves your library; the side panel shows the UI. REQUIREMENTS Chrome 114+ (side panel). Stash works on standard http(s) pages. Built-in pages (chrome://), the Chrome Web Store, PDF viewer tabs, and some restricted URLs cannot be scripted by policy—open a regular webpage to test. WHO IT'S FOR Researchers, writers, developers, and anyone who wants durable, portable notes from the web without another account. WHAT USERS CAN DO - Turn the current tab into Markdown: stash articles, docs, and PDFs from Chrome’s side panel. - Drop local documents straight in: PDF, Word (.docx), PowerPoint (.pptx), and Excel (.xlsx / .xlsm) files become searchable Markdown entries. - Keep everything local: conversion runs in the browser from the active tab or local file blob; no cloud converter, sync backend, or analytics endpoint. - Build a portable research library: organize by groups, search, batch move/delete/download, copy Markdown, export JSON backups, combined Markdown, or llms.txt. - Make spreadsheets useful for AI workflows: Excel sheets are rendered with explicit cell coordinates (A1=value) plus detected table regions, so pricing tables and messy workbooks keep their structure. Changelog v1.1.3 — Local file drag-and-drop ingestion (2026-05-05) Drag-and-drop ingestion: drop local PDFs, Word (.docx), PowerPoint (.pptx), and Excel (.xlsx / .xlsm) files directly onto the side panel. Modular local ingestors: added office-ingest.js with separate adapters for PDF, DOCX, PPTX, and Excel OpenXML so each extraction technique can be replaced independently. Excel extraction: ports the local workbook parsing shape from excel-pricing-tables: visible sheet cells are normalized, trailing empty cells are trimmed, rendered as A1=value, and locally detected table regions are included. Privacy posture unchanged: Office and PDF imports are parsed in the browser from dropped/picked file blobs. No cloud conversion, no upload, no remote service. Package: dist/mdream-stash-extension-chrome-store-v1.1.3.zip from manifest.json version (run scripts/package-extension.ps1 or .sh). v1.1.2 — Side panel redesign (2026-05-05) Visual refresh: darker “vault” shell, warm paper cards, amber accents, stronger hierarchy in header and list. Group filters: horizontal scrollable filter rail instead of a wrapping chip grid. Library rhythm: list grouped by recency (for example Today / Yesterday) with short Markdown previews on each card. Store listing: after upgrading, regenerate screenshots so the Chrome Web Store matches the new UI — npm run capture-screenshots (see docs/store-assets-plan.md). Package: dist/mdream-stash-extension-chrome-store-v1.1.2.zip from manifest.json version (run scripts/package-extension.ps1 or .sh). v1.1.1 — Data residency checks & packaging (2026-04-13) Automated data residency policy: static scan (npm run test:policy) plus a Playwright network audit so idle side panel + PDF WASM init never open remote http(s) URLs from extension contexts. See [docs/data-residency-policy.md](docs/data-residency-policy.md). Imported PDF refresh: clear error instead of a broken fetch when refreshing an imported PDF (no URL). Store zip filename: packaging scripts now write dist/mdream-stash-extension-chrome-store-v<version>.zip using manifest.json version. v1.1.0 — PDF Support (2026-04-13) What's new for users: Stash PDFs directly from the browser. Open any PDF in a Chrome tab and click "Stash This Page" — the extension now detects PDFs automatically (by URL or content-type) and extracts structured Markdown with headings, tables, lists, and multi-column reading order preserved. Import local PDF files. Use Import (arrow-up button) and pick a .pdf file from your disk. The text is extracted entirely on your device — nothing is uploaded anywhere. PDF badge in the item list. PDF items show a red "PDF" badge instead of a website favicon so you can tell them apart at a glance. Refresh works for PDFs too. Re-stash a PDF item to pick up changes without opening a new tab. Better error messages. If a PDF is image-only or corrupted, you get a clear explanation instead of a generic failure.