audiobook-download

name

audiobook-download

description

Finds and downloads free, legally available audiobooks — public-domain recordings from LibriVox and the Internet Archive — via curl + jq, no API key needed. Saves into /Volumes/SanDisk/Audiobooks/ (syncs to phone) when mounted, else the current topic folder, and sends the file(s) via the bot MCP `send_document`. Does not touch piracy sites that host copyrighted commercial audiobooks without a license — that's out of scope. Use when: "скачай аудиокнигу", "найди бесплатную аудиокнигу", "скачай книгу в аудио", "где послушать книгу бесплатно", "хочу аудиокнигу", "download an audiobook", "find a free audiobook", "get me the audio version of [book]"

Audiobook Download

Downloads audiobooks that are actually free to distribute: public-domain works read by volunteers (LibriVox) or hosted with permission on the Internet Archive. Both are queryable with plain curl + jq — no account, no key.

This skill will not scrape sites that redistribute copyrighted commercial audiobooks without a license — that's piracy, not "free," and creates legal exposure for whoever hosts and shares the file. If a requested book is a modern bestseller still under copyright, read legal-sources.md for library-lending and other legitimate free routes to offer instead.

Workflow

1. Search the Internet Archive. LibriVox uploads every recording to archive.org already split into chapter-level MP3s, so one API covers search + file listing:

   curl -s "https://archive.org/advancedsearch.php?q=title:(BOOK+TITLE)+AND+collection:(librivoxaudio)&fl[]=identifier&fl[]=title&fl[]=creator&fl[]=downloads&rows=10&output=json" \
     | jq '.response.docs'
   

   No hits in librivoxaudio? Broaden to any public-domain audio item:

   curl -s "https://archive.org/advancedsearch.php?q=title:(BOOK+TITLE)+AND+mediatype:(audio)&fl[]=identifier&fl[]=title&fl[]=creator&fl[]=downloads&rows=10&output=json" \
     | jq '.response.docs'
   

   Pick the best match by title/creator; if several editions/readers show up, prefer the one with the highest downloads and confirm with the user if it's ambiguous.

2. List the item's files with the Metadata API:

   curl -s "https://archive.org/metadata/${IDENTIFIER}" \
     | jq '.files[] | select(.name | test("_64kb\\.mp3$|\\.m4b$")) | {name, format, size}'
   

   You'll typically see a single-file .m4b (whole book, chapters embedded) and a set of per-chapter _64kb.mp3 files (smallest bitrate offered — pick these over the plain/_128kb MP3s or .ogg).

3. Download. Telegram documents cap around 50 MB — check size first. The .m4b is usually way over that for a full-length book, so default to per-chapter 64kbps MP3s:

   curl -sL "https://archive.org/download/${IDENTIFIER}/${FILENAME}" -o "${FILENAME}"
   

   Where to save: if /Volumes/SanDisk/Audiobooks/ is mounted (check with ls first), save there in a Автор - Название/ subfolder, matching the existing library's naming convention — that drive syncs to the user's phone. If it's not mounted, fall back to the current topic folder and say so, so the user knows to move the files later.

4. Send the file(s) via the bot MCP send_document, one message per file, regardless of where they were saved — so the user has an immediate copy in chat even before the drive syncs. For multi-chapter books, send a short caption first (title, author, reader, chapter count) so the user knows what's coming, then the files.

5. Not found anywhere on archive.org (usually means it's still under copyright, e.g. a recent bestseller) → tell the user directly that there's no legal free copy, and offer the library-lending and other legitimate options from legal-sources.md instead.

Notes

• No auth, no rate-limit headers documented for either endpoint — keep request volume reasonable (a handful of calls per search, not loops).
• advancedsearch.php treats spaces in q as AND between terms inside the parens; wrap multi-word titles/authors in parens as shown above.
• Non-English public-domain audiobooks exist on archive.org too, but coverage is thin outside English — a creator:(Tolstoy) search mostly returns English/French/Spanish translations, with only a handful of actual Russian-language recordings. Check the item's language/title before downloading, and set expectations with the user accordingly: classic authors dead 70+ years may have a Russian recording, but often don't, and living/modern authors won't have one anywhere legally.
• If the user just wants to browse rather than have Claude fetch a specific title, legal-sources.md also lists sites worth pointing them to directly.