updates

FlowPick v1.0.0 Release Notes: The First Public Drop

FlowPick v1.0.0 is live — a browser extension that auto-detects HLS/DASH streams, audio, and images, merges fragments in-browser via WebAssembly FFmpeg, and keeps everything local. Here's what shipped on day one.
FlowPick Team
8 min read
# release # v1.0.0 # launch # changelog # webassembly # ffmpeg

After three months of late nights, broken WebAssembly builds, and one near-fatal decision to rewrite the fragment merger in Rust before chickening out, FlowPick v1.0.0 is out. This is the first public release — the one where we stop pretending it's a "preview" and let actual humans install it.

Download v1.0.0 from GitHub Releases →

The pitch is simple: a browser extension that sniffs M3U8/HLS, MPD/DASH, MP4, audio, and images — then merges the fragments in-browser using WebAssembly FFmpeg, without a desktop app, without a server, without anything leaving your machine. If you've ever wanted to grab a stream from a niche site and been told "just install yt-dlp and FFmpeg and configure your PATH," this is the middle finger to that workflow.

Here's what actually shipped in v1.0.0.

Video sniffing — the core

The headline feature. FlowPick watches the browser's network requests and picks out streaming manifests before you even think about opening DevTools.

  • M3U8/HLS auto-detection. Adaptive bitrate streams get parsed on the fly; you see every quality tier (1080p, 720p, etc.) and pick the one you want. The HLS/M3U8 download guide walks through how this actually works under the hood.
  • MPD/DASH support. Same deal for DASH streams — audio and video tracks are detected separately and merged during download. If you're curious why DASH splits audio and video in the first place, the DASH/MPD explainer covers it.
  • In-browser fragment merging. This is the magic. Hundreds of .ts or .m4s segments get concatenated and muxed into a single MP4 using WebAssembly FFmpeg — no desktop FFmpeg, no ffmpeg.exe in your PATH, nothing. The technical writeup on in-browser merging goes deep on the implementation.
  • Direct MP4/WebM/FLV download. For sites that serve plain video files instead of streaming, FlowPick just grabs the URL.

Audio capture

Audio resources get their own detection pipeline — MP3, AAC, WAV, OGG, FLAC, M4A. There's a built-in player so you can preview before committing to a download. Podcasts, background music, random sound clips — if the browser loads it, FlowPick sees it. The audio sniffing tutorial covers the workflow.

Image batch download

This one surprised us — people use it more than expected. Batch-download every image on a page (JPG, PNG, GIF, WebP, SVG, AVIF), filter by size/format/dimensions, skip the thumbnails. Pinterest boards, product galleries, wallpaper sites — the batch image download guide has the details.

Smart filtering

Once you've sniffed a media-heavy page, you need to actually find what you want. Filters:

  • By type — video, audio, image
  • By size — auto-skips ad segments (those tiny 2-second .ts files that pollute HLS streams)
  • By duration — for video and audio
  • Search — within detected resources

Download manager

  • Multi-threaded parallel fragment downloads
  • Progress tracking with ETA
  • Batch queue
  • One-click "Pick" button on any resource

Privacy & security

This is the hill we'll die on. FlowPick collects zero data. No analytics, no tracking, no "anonymous usage statistics." Everything happens locally in the browser sandbox. No server ever sees your URLs, your cookies, or your downloaded files. Read the full privacy & security docs if you want the receipts.

  • Zero data collection
  • 100% local processing
  • Dark mode (follows system preference)
  • Keyboard shortcuts: Alt+Shift+F to open the sniffer, Alt+Shift+D for the download manager

Browser support

BrowserStatusNotes
Google Chrome✅ SupportedRecommended
Microsoft Edge✅ SupportedChromium-based
Mozilla Firefox✅ SupportedFull feature parity

Firefox support shipped same-day as Chromium — no second-class citizen treatment here.

What's next

v1.0.0 is the foundation. Future versions will bring format conversion options, advanced filtering rules, more languages, and performance work. The v1.1.0 release notes are already up — that one adds MSE recording, document sniffing, and the grouped video card UI.

If you hit a bug, file it on GitHub. If you want to follow updates, the changelog page has every version.