AI Readers

First time with AI Readers

• 01

  AI Beta Readers vs Slima AI Editor

  AI Beta Readers simulate readers to give you reader-experience feedback; Slima AI Editor acts like an editor doing a 5-dimension pass. Different purposes, complementary.

• 02

  Four-step flow & privacy

  Beta Readers follows four steps: pick book → pick service → pick personas / dimensions → run report. How your content is handled and not used for training.

• 03

  Upload a manuscript or pick from Slima

  Two entry paths to Beta Readers: pick from your existing Slima books, or upload an external manuscript (Word / Markdown / Scrivener / plain text).

AI Readers list

• 01

  AI Readers list overview

  Entering AI Readers shows your reading history — sorted by time, each row shows book, service, status (running / done / failed), and report link.

• 02

  Trash & restore

  Deleted readings land in the list's Trash tab, recoverable for 30 days. After that, permanent (including report content).

Set up a reading

• 01

  Set up a reading: workspace layout

  The set-up screen has four regions: book preview, service choice, config (personas / dimensions), confirm + cost estimate.

• 02

  Select content: which chapters get read

  Slima reads every chapter by default; you can include / exclude specific ones (e.g. only first 3). Script Studio adds scene-level selection.

• 03

  Pick service: Beta Readers or Editor

  Pick which service: AI Beta Readers (simulated readers) or Slima AI Editor (editor's eye). Differences and how to choose.

• 04

  Pick test type: full vs quick

  Pick the run intensity: full (reads everything, all sections) or quick (reads a sample, returns 1–2 key sections).

Reader personas

• 01

  Personas library: attributes and picking

  Slima ships a persona library — each persona has age, reading preference, cultural background, tolerance attributes. How to pick; can you customise.

• 02

  Filter / select / clear personas

  Persona picker operations: filter by attribute, check / uncheck, save combo (Pro+), clear.

• 03

  Persona simultaneous limit and upgrade prompt

  Free / Starter / Pro have different simultaneous persona limits. Hitting the cap pops AiGateModal explaining + offering upgrade.

Run a reading

• 01

  Start, progress, cancel

  Hit "Start" and Slima runs the reading in the background — progress bar shows; you can leave. Cancel mid-flight keeps what's been produced.

Report structure

• 01

  Report structure: overview + 5 tabs

  A full reading report has five tabs: overview, reader feedback, editor letter, scoring, action items. Each emphasises different things.

• 02

  Dashboard four metrics

  The overview at the top of every report shows four metrics: overall score (0-10), Kindle viability (high / medium / low), pacing score, completion rate.

Overview dashboard

• 01

  Reader feedback tab

  Main report tab — overall prediction + 7 diagnostic sections (opening / characters / pacing / context / market / ending / arc) + individual reader cards.

Reader feedback overview

• 01

  Overall + Kindle viability prediction

  Top-of-report overall prediction: completion rate, drop-chapter, emotion distribution; Kindle viability is a composite prediction of "can it hit category top 100 week one".

Seven diagnostic sections

• 01

  Opening diagnostic

  The opening diagnostic scores whether the first chapter / first 1000 words hook readers. Drops happen earliest at the opening — this section is high-priority.

• 02

  Characters diagnostic

  Characters diagnostic scores protagonist likeability, supporting-cast depth, arc completion, voice consistency. Often pairs with "emotional investment low" reader feedback.

• 03

  Pacing diagnostic

  The pacing diagnostic scores each chapter and assembles a heatmap. You see at a glance which chapters drag and which rush.

• 04

  Context and market diagnostic

  Context: is worldbuilding / background set up clearly. Market: how does this compare to peers — what's familiar, what's distinctive.

Individual reader cards

• 01

  Individual reader cards

  One card per persona: where they dropped, their most-liked / most-confused passages, aggregate emotion, a 3-sentence review. Concretises abstract reader feedback.

Editor letter

• 01

  Editor letter tab

  Slima AI Editor's overall take, written as a letter (not a bullet list) — read in one pass for an overall impression.

Editor five lenses

• 01

  Editor's 5-dimension pass

  The core of AI Editor service: five independent dimensions scoring the book — structure, characters, language, pacing, market. Per dimension: scores + cited passages + suggestions.

• 02

  Severity tags and QA block

  Every AI Editor suggestion has a severity tag (must-fix / recommended / optional). A QA block at the bottom lets you ask the editor follow-ups.

Action items

• 01

  Action items and one-click AI prompts

  The "Action items" tab consolidates must-fix and recommended suggestions. Each has a "send to AI Coach" button — opens a new chat with context pre-loaded.

Rating scale

• 01

  Scoring criteria

  How Slima derives the 0–10 scores. What goes into each dimension, why it's defined that way, why scores are directional not absolute.

Reading history

• 01

  Reading history & revisit

  Every reading persists in history. Open old reports, compare scores over time, diff action item lists.

Report quota & cost

• 01

  Credits estimate and quota display

  Set-up shows a live "this run needs X credits, you have Y" estimate. Credit consumption depends on chapter count, persona count, service type.

• 02

  Word limit and upgrade prompt

  Each tier has a per-run word ceiling. Hitting it pops AiLimitReached + upgrade prompt. Free only reads short pieces; Pro+ reads full long books.

Export report

• 01

  Export report — Markdown / Word

  The report's top-right "Export" sends the whole report / individual tabs as Markdown / Word. For archiving, sharing with editors, posting elsewhere.