Community · E2 · artifact verified
Tag a calibre library with typed suggestions
A calibre plugin that suggests subject and genre tags for your books with Jev: suggestions arrive for review before anything is applied, existing tags are preserved, and the plugin is independent GPL software with a live demo and production notes.
01 · Role in the system
What Jev does here
For each book the plugin sends its metadata as state with typed questions to api.typesafe.ai's systemone endpoint and gets back suggested subject and genre tags as structured answers rather than a paragraph to parse. Nothing is applied automatically: suggestions queue for review, existing tags are always preserved, and the library's own category vocabulary can be imported so suggestions land in terms you actually use. It ships as a normal calibre plugin (GPL-3.0-or-later, explicitly unaffiliated), with a demo gif, a live presentation site, and a production-notes document recording full results from real runs.
02 · Control boundary
Where Jev sits
Jev as the tag suggester inside an existing library manager: typed questions over book metadata, structured suggestions queued for human review, imported category vocabularies keeping the answers inside the taxonomy you already use.
Code owns the loop, permissions, thresholds, validation, and side effects. Jev owns only the bounded judgments described above.
03 · Known limits
What this evidence does not prove
- Suggestions depend on metadata quality; books with sparse records get weaker tags.
- New project (created 2026-10-01, 0 stars); results documented by the author, not independently evaluated.
- Requires your own TypeSafe API key; per-call charges apply.
04 · Attribution
Public sources
This is a Community record: the project was published by a third-party community author.
- iamjonatha ↗Community · github · public · checked 2026-10-01