Skip to main content

​
Calibre-Web Automated

Ebook library management with auto-ingest, format conversion, OPDS catalog, and KOReader sync. Replaces Boundless. Runs on i3 (debian-1) as a Swarm service (calibre-web-automated_calibre-web-automated) on the swarm overlay. Image crocodilestick/calibre-web-automated:latest, port 8083. The earlier “runs on Pentium” description is obsolete.

​
URL

ServiceURL
CWAhttps://calibre.augustin.ai
OPDShttps://calibre.augustin.ai/opds

​
How It Works

Chaptarr downloads ebook → custom script copies to CWA ingest → auto-ingest into Calibre
  → auto-convert to epub
  → Hardcover metadata enrichment
  → OPDS catalog for KOReader
  → KOSync reading progress sync

​
Enabled Features

  • Auto-ingest from /cwa-book-ingest watch folder
  • Auto-convert to epub (28 formats supported)
  • OPDS catalog at /opds (authenticated)
  • KOReader sync — reading progress across devices
  • Kobo sync protocol
  • Hardcover sync — metadata enrichment from Hardcover API
  • Kindle EPUB fixer — fixes common epub issues
  • Duplicate detection — smart dedup on ingest

​
Architecture

calibre.augustin.ai → Cloudflare tunnel → Traefik (i3, Swarm provider) → calibre-web-automated (overlay)
Routed by Traefik’s Swarm provider via deploy.labels in calibre-web-automated/compose.yaml.

​
Chaptarr Integration

Ebooks reach CWA’s ingest folder via host cron jobs on i3, not via Chaptarr’s custom script. The in-container custom script (/config/scripts/cwa-ingest.sh) is wired but does not fire on real imports — only Test events ever reach it (Chaptarr notification bug). The working path is downloads/cwa-ingest-cron.sh / downloads/torrent-to-cwa.sh. See the Reading Pipeline doc for the full (and currently redundant) ingest story.