A CLI (Deno/TypeScript) that downloads an entire Skool classroom: all videos (Skool-native Mux + embedded providers) and lesson attachments, organized into a structured folder tree on disk.

Disclaimer: This tool is meant for archiving content you have legitimate (paid) access to — e.g. making your own courses available offline. Respect the Skool Terms of Service and the copyright of the creators. Do not download or redistribute content you do not have the rights to. Use at your own risk.

• Deno ≥ 2.8
• yt-dlp and ffmpeg in your PATH
• cookies.txt (Netscape format) exported from a logged-in browser

Run a browser extension like "Get cookies.txt LOCALLY" on www.skool.com and save the result as ./cookies.txt (must contain auth_token). The cookie expires — re-export it on auth errors. See cookies.example.txt for the format.

deno task dl <url> [options]

Examples (replace <your-community> with your community slug):

# whole classroom (dry-run first)
deno task dl "https://www.skool.com/<your-community>/classroom" --dry-run
deno task dl "https://www.skool.com/<your-community>/classroom" --quality 720

# single lesson
deno task dl "https://www.skool.com/<your-community>/classroom/<root>?md=<lessonId>" --lesson-only

downloads/<course>/<module>/NN-<lesson>/video.mp4

Already-downloaded files are skipped (resumable).

• Skool is built on Next.js; page data lives in <script id="__NEXT_DATA__">.
• Classroom tree: renderData.allCourses[] → per course pp.course (modules/lessons).
• Per lesson, /<slug>/classroom/<root>?md=<lessonId> returns the Mux playbackId + signed playbackToken (JWT) in pp.video. Stream: https://stream.mux.com/<id>.m3u8?token=<jwt>, downloaded via yt-dlp with Referer: https://www.skool.com/.

deno task test

metadata.resources is a JSON string containing {title, file_id, file_name, file_content_type}. The actual download link is resolved via POST https://api.skool.com/files/{file_id}/download-url (a signed files.skool.com URL) and then fetched.

• deno run --allow-read --allow-net src/diagnose.ts <slug> [slug...] — finds lessons that should have a video but where the download failed.
• deno run --allow-read --allow-net src/reconcile.ts <slug> [slug...] --out <dir> — lists videos that exist according to the crawl but are missing as .mp4 on disk.

• Embeds (Loom/Vimeo/YouTube/Wistia) are detected from videoLink + desc.
• Sequential download (to be gentle on rate limits). No posts/comments (YAGNI).

MIT — see LICENSE.