Skip to main content

1,104 AI-written novels, free to read: the Dracon Bookstore and how it was made

The Dracon Bookstore holds 1,104 novels, 131,478,779 words across 22,064 chapters, all written end to end by our own book pipeline. No crowdfunding, no soft launch, no shrug. The books say how they were made, they are free to read, and they always will be. Here is the shelf, then the honest part: how the engine works and where the label lives.

We opened the Bookstore on August 1, 2026 with five full novels. Those five have since been regenerated out of the catalogue, because the pipeline re-draws the shelf as it improves, and the books in the table below are today's flagships instead. The engine, the label and the promise have not changed. The shelf now holds **1,104 books, 131,478,779 words across 22,064 chapters**, under 1,085 pen names.

The shelf

Today's flagship titles from web/books/src/lib/data/catalog.json, the catalogue snapshot of September 24, 2026. The whole shelf is at /books.
TitleBylineNicheGenreWordsChapters
The Soil Between UsElena VorneSlow burnRomance136,97120
The Seventh ConfessionAdair PageCold caseMystery136,61720
The UnspoolingVane RookSpace operaSci-Fi135,52920
An Oath of CindersCade VossSlow burnFantasy132,93420
The Wrecking LightSoren StarlingLiterary fictionHorror129,37220

The spread is deliberate. The Soil Between Us is a slow burn romance, The Seventh Confession is a cold case mystery, The Unspooling is a space opera, An Oath of Cinders is a slow burn fantasy and The Wrecking Light is literary fiction with a horror spine. Five genres, five pen names, one engine. The shelf is far bigger than five now, and the fastest way into it is not a table like this one.

Browse by trope

A reader does not arrive wanting a genre. They arrive wanting the feeling: enemies to lovers, a slow burn, a story where nobody is a hero, a book with no heat in it. So the shelf is cut by trope as well, and every one of those collections is a real page with its own shelf, its own questions-readers-actually-ask block and its own free samples. The biggest right now are enemies to lovers, slow burn, morally gray, romantasy and no spice, which is the 1,027-book set of everything that stays closed door. All 61 trait shelves are here. The 7 genre shelves sit beside them for a reader who thinks in shelves rather than feelings: romance, mystery, fantasy, horror, scifi, fiction and nonfiction.

Each book page carries its own traits, and each trait carries a shelf, so you can go in from the feeling and come out on a book. The books are free to start: the first fifth of every one of them opens in the browser, with no account and no card.

How the books were made: pipeline, not prompt

The shelf data carries the receipt. Every book names its source, and the source is our own pipeline, not a text generator with a public API behind it. Bylines are pen names for pipeline configurations, not people: the catalogue now carries 1,085 of them, each mapping to a genre and pacing preset. The pipeline runs in stages and emits complete chapter files, not a chat log stitched together.

  • Premise and bible: genre, blurb, discovery metadata (tropes, mood, pacing, heat, themes), and a 20-chapter outline are generated and frozen. The outline is the contract: chapters cannot invent a protagonist in chapter 14.
  • Chapter drafting: each chapter is generated against the bible plus prior summaries, 4,000 to 7,000 words. The JSON records per-chapter counts so length is auditable.
  • Continuity pass: names, timelines, and clues are checked, so a planted detail in chapter three still lines up in chapter nineteen.
  • Rollup: 1,104 books, 131,478,779 words across 22,064 chapters, same source tag and same pipeline. Every book is a 20-chapter manuscript of roughly 119,000 words.

The cookbook: recipes, not magic

Each novel is built from a cookbook of versioned prompt recipes (premise, outline, chapter, revision) with style constraints. Like the cooks in our music library, a recipe turns ingredients into a track; a book cook turns a genre and premise into a manuscript. The recipes are the IP, the manuscripts the demo, and the future member tool the kitchen where you point the engine at your premise.

  • Premise cook: blurb plus tropes, mood, pacing, heat and themes.
  • Outline cook: 20-chapter beats from premise and bible.
  • Chapter cook: 4,000 to 7,000 word chapter from outline, bible and prior summaries.
  • Continuity cook: flagged inconsistencies and patches.
  • Cover prompt cook: image prompt for hand-painted cover style, no text on cover.

The ethics disclosure on the tin

Readers forgive a machine-written book; they do not forgive being tricked into one. The shelf, each book page, each free sample, each chapter and every collection page carry the same disclosure: these are AI-pipeline novels, the byline is a configuration, and the source is our own pipeline. Open any book at /books and the label is on the page itself, next to the byline, not a JavaScript footnote.

  • We never claim a human ghostwriter. The byline is the pen name, provenance is the pipeline, chapters are generated, not dictated or lightly edited from a human draft.
  • We never paywall the free sample. Every book opens its first fifth in the browser at /books and on its own page, and the books are free to read forever. The bookstore is the demo, not the product.
  • We never train on your reading. Opening a chapter does not feed a model; highlights and likes stay in the product domain, not a training set.
  • We label growth. A book without the label is not ours, and a new title inherits the tag from the pipeline that made it.

Who should read these, and who should not

If you want to judge whether a pipeline can sustain a novel (chapter structure, planted clues, a protagonist who changes), read chapter one then skip to chapter ten to see if the thread held. The Seventh Confession is the best test for clue discipline, The Unspooling for pacing under fire, and An Oath of Cinders for slow-burn atmosphere. If you want a human memoir voice or research-heavy nonfiction, these are not that. Better you know in paragraph one than chapter twelve.

Limitations are part of the label. Pipeline prose can be even when it should be rough, tidy when it should be tangled, generous with description when a scene needs to breathe less. We edit for continuity, not for voice transfusion: the pipeline's cadence is the book's cadence. The future member controls will expose heat, pacing, and mood as dials; today they are fixed labels on the shelf.

Where to start, and what comes next

Start on the shelf by itch, or on a trope shelf if you know the feeling you want. Slow and sure: The Soil Between Us. Cold and procedural: The Seventh Confession. Big and structural: The Unspooling. Bleak and beautiful: The Wrecking Light. Every page lists words, chapters and tropes. When the member tool ships it will live alongside these free proofs (same pipeline, your premise). Until then the books are evidence, free forever, with the AI label where it belongs.

Sources & provenance