Book Aurora
Visualizer that sends every passage of a novel to Jev for nine emotion scores plus intensity and draws the book as an aurora poster.
Overview
Book Aurora has Jev read a whole novel and turns each passage into a row of colour. Each roughly 90-word passage gets ten parallel typed questions: a 0 to 3 score for each of nine emotions, plus overall intensity. Jev never sees pixels and never writes prose; the app draws one feathered row per passage with each emotion as a curtain whose width and brightness follow its score, and the finished strip can be saved as a poster.
The README reports Frankenstein as 601 passages and 6,010 typed decisions in about 22 seconds without screen recording, for about 3 cents. It runs with Bun and Vite, calls Jev through OpenRouter's decisions endpoint or the native API, and falls back to a keyword mock mode for visual work. Options include concurrency, timeouts and request hedging (a duplicate request fires if the first is slow).
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Score | How strongly does this passage express each of nine emotions? | 0 · 1 · 2 · 3 |
| Score | How intense is this passage overall? | — |
How it works
- Split the book into ~90-word passages
- Send ten score questions per passage
- Draw a row of colour from the scores
- Export the full strip as a poster
Strengths
- Whole novel in seconds for cents, per the README
- Hedged requests reduce tail latency
- Mock mode for development
Caveats
- Visual art piece, not literary analysis
- Book text is sent to the API
Getting started
bun install && bun run dev, with OPENROUTER_API_KEY or TYPESAFE_AI_API_KEY set.At a glance
- Type
- Demo
- Status
- prototype
- Language
- TypeScript
- Runtime
- Bun API server + Vite web app
- Has code
- yes
Use cases and integrations
How curated lists describe it
Jev reads a whole novel in seconds: each passage gets nine emotion scores plus intensity in one call, and every passage becomes a feathered row of colour. Frankenstein is 601 passages, 6,010 typed decisions, about 25 s and 3 cents; exports a poster.
AnotiaWang/awesome-decision-models
Frequently asked questions
What does Jev score?
Nine emotions on a 0-3 scale plus overall intensity per passage.
How much does a book cost?
Frankenstein cost about 3 cents, per the README.
Can I run it without a key?
Yes, in a keyword mock mode for visuals.
Related projects
Interactive art piece where Jev reads a post as 100 fictional observers, picking an intuition and scoring four feelings for each.
ProjectbetaTypeScriptEmoji autocomplete that asks Jev eight parallel questions per typing pause: emoji choice, emotion, mood, urgency, sarcasm and more.
ServiceprototypeTypeScriptUndocumented app that uses Jev to classify text and pick an emoji reaction.
ProjectprototypePythonCompanion that answers only with facial expressions: Jev picks one of 16 moods for each message and a blob avatar morphs its face.
ProjectprototypeTypeScriptWeb demo where you paste text and Jev suggests the best-fitting emoji.
DemoprototypeCLI that has Jev judge a video's subtitles segment by segment, then cuts a highlight reel and a filler-free full version.
ProjectbetaPython