Keep the web pages you care about.
WebFrost archives a URL to every format worth keeping, on your own machine, and lets you search what you saved. Run it from a terminal, a script or a browser.

One URL in, every format out.
Start the server, or just use the CLI.
The Docker image bundles Chromium, yt-dlp and ffmpeg, so every format works out of the box. The same archive folder works from the CLI and from code.
curl -fsSLO https://webfrost.dev/compose.ymlecho "SESSION_SECRET=$(openssl rand -hex 32)" > .envdocker compose up -d# open http://localhost:3000 and create the admin userArchive once, keep it in every shape.
Every format from one visit
One fetch and one browser navigation produce WARC, raw HTML, readable text, PDF, a full-page screenshot, the rendered DOM, a self-contained HTML file and, for media pages, the video.
Keeps running on its own
Cron schedules and RSS feeds run inside the server. Jobs are queued with a concurrency limit, retried after a crash, and spaced out per site so nobody gets hammered.
Safe to put on a network
Sessions you can revoke, API keys, rate limits, and a guard that stops archived pages — and the browser rendering them — from reaching your internal network.
Plain files, no lock-in
An archive is a folder: one SQLite database and the files next to it, or an S3-compatible bucket. Copy it, back it up, read it with any tool.
Use it the way that fits.
Share an archive without sharing the keys
Public read mode lets anyone browse and search; only you can archive, delete or change settings.
