kosal-link

Scrape any landing page from your terminal.

A CLI that captures JS-rendered pages locally with Playwright. Sign in once with GitHub and every scrape runs on your machine — no servers, no uploads.

Install GitHub

Install

Requires Node 18+. Chromium is installed automatically.

npm install -g kosal-link

Or with pnpm / yarn:

pnpm add -g kosal-link
yarn global add kosal-link

Login

Authenticate with GitHub. A browser window opens and a token is saved to ~/.config/kosal-link/config.json.

kosal-link login

Scrape a page

Paste any URL. Output is saved to ./<hostname>/ by default.

kosal-link scrape https://example.com

Specify an output directory:

kosal-link scrape https://example.com -o ./my-site

Output structure:

./example.com/
├── index.html         # hydrated HTML with local asset paths
├── screenshot.png     # full-page screenshot
└── assets/            # every JS/CSS/image/font captured

Other commands

kosal-link whoami    # show the signed-in GitHub user
kosal-link logout    # revoke token and clear local config
kosal-link --help    # full command list

Options

Flag	Default	Description
`-o, --output`	`./<hostname>`	Output directory
`--timeout`	`45`	Navigation timeout (seconds)
`--no-screenshot`	off	Skip full-page screenshot