Labs · Novel to lorebook
The lorebook’s skeleton,
counted out of the book itself.
Drop a txt or epub: chapters get detected, recurring characters, places and terms get counted by rule, and out comes a SillyTavern World Info skeleton — every entry with its trigger key, chapter range and token budget, waiting for you (or an AI with the included brief) to write the flesh. Zero inference, zero invented lore, nothing uploaded.
Drop a novel here, or click to choose
.txt / .epub · 30MB tomes parse locally
Parsing, counting and skeleton building all run in your browser — the book is never uploaded or stored.
Drop a book — the entity list and skeleton appear here.
Where this tool comes from
The chapter detector is the same production-ported engine behind our Book Checkup (第N章 families, numeric headings, English chapters), and the exported JSON uses the ST 1.18 entry template — the same field family our Lorebook Merger writes, so it imports without translation. The entity rules were tuned on Chinese web fiction: speaker verbs are the strongest person signal a rulebook can have, suffix patterns catch places and sects, quoted brackets catch techniques and artifacts. Frequency thresholds (2 speaker hits, 3 otherwise) and a 60-entity cap keep the output a skeleton, not a noise dump.
What it can’t do (the honest part)
It doesn’t understand the story — it counts it. A character who never speaks, a place named without a suffix, an alias the book uses more than the real name: all can slip through, and the table says what evidence each entity rests on. Entry bodies are templates on purpose: filling them with real lore requires reading the book, which is exactly the step we hand to you or your AI via the brief. And chapter ranges are only as good as chapter detection — when a book defeats the detector, the page warns you instead of pretending.
FAQ
Does it write the lorebook entries for me?
No — deliberately. Entry bodies need someone who actually read the book: this page runs zero AI and invents zero lore. What it does produce is everything mechanical about a lorebook: which entities exist, their trigger keys, which chapters they span, how often they appear, and a token-budget hint per entry. The fill-in brief it copies out carries the craft rules, so any AI (or the Foreverse in-app butler agent) can finish the job against the source text.
How are entities found without AI?
Rule signals, counted honestly. Chinese books: speaker patterns (a 2–4 character name directly before 说道/问道/笑道…), place and organization suffixes (城/镇/山/宗/学院…with junk characters stripped from the left), and quoted terms (「青冥诀」). English books: capitalized phrases that appear mid-sentence, with a speaker-verb check for characters. Everything below the frequency thresholds (2 hits for speakers, 3 for the rest) is dropped — the tool prefers missing an entity to inventing one.
What do I do with the JSON it exports?
It's a standard SillyTavern World Info file — the same field family ST 1.18 writes. Import it into ST's World Info panel (or Foreverse), and each skeleton entry appears with its trigger keys, ready for you to replace the template text with real lore. Entries you didn't check in the table stay out of the file.
Why did it miss a character I know is important?
Three usual reasons: the name never appears in a speaker pattern (a silent character), the book's chaptering defeated detection (ranges then default to chapter 1), or the name is below the frequency threshold. That's the honest cost of rule extraction — add the missing entity as one more entry in the exported JSON; the format is plain to edit.
Is my book uploaded?
No. txt decoding, epub unpacking, chapter detection, entity counting and JSON building all run in your browser. Works offline once loaded.