Capcat – CLI/TUI to Archive Articles as Markdown and HTML (FOSS)

Made this for myself first. I read a lot, save things in multiple places, then can't find them later, or the page is just gone. So I wanted one quick archive (in any folder of my choosing): plain Markdown on disk, with images and attachments, organized by date and source.

Two ways to use it: a CLI for scripting and cron, and a TUI for when I don't feel like typing. Same backend, identical output. In the CLI, add --html and you get self-contained themeable pages - dark/light, fully offline, and editing one CSS file restyles the entire archive. The same options are exposed in the TUI. The Markdown drops straight into the folder, so you can load it in Obsidian easily (front matter included).

For sources it leans on official APIs and RSS before it ever scrapes HTML. Uses HN's official API, respects robots.txt, rate-limits itself to 1 request / 10s, names its user agent honestly, skips paywalls. Each source has its own YAML config, and the configs and themes live inside the archive folder itself, so the whole archive is one self-contained, portable directory. (If you use Obsidian, that folder is just your vault.)

Repo:https://github.com/stayukasabov/capcat.

Tear it apart, feedback welcome.