AI Readers

AI Beta Readers + Slima AI Editor: reading reports & 5-dim editorial pass

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 / plain text / PDF / EPUB).

AI Readers list

• 01 AI Readers home: your works AI Beta Readers' home is a shelf of your works (scripts + books), each showing type, sync status, and readings count. Click a work to set up a reading; each work has a Reading History tab; deleted runs go to Trash.
• 02 Delete a reading Deleting a reading removes it from your Reading History list. There's no restorable trash view yet, so confirm before you delete.

Set up a reading

• 01 Set up a reading: workspace layout The set-up screen has two columns: a left panel (content / history) and a center column that stacks selected content, service choice, config (personas / dimensions), and the report-quota summary with a Start button.
• 02 Select content: which chapters get read Slima reads every chapter by default; you can include / exclude specific ones (e.g. only first 3). For a script, the whole script is auto-assembled and read as one unit.
• 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 Set up a reading: content, readers, and scope Set up a reading on the set-up screen: choose which content to include, pick your readers and/or editor analysis, then press Start.

Reader personas

• 01 Personas library: the built-in AI Beta Readers Slima ships 8 fixed AI Beta Readers, each a critic lens (plot holes, hard-to-impress, emotion, lore, prose, speed, mainstream, foreshadowing). How to pick a combination for your book.
• 02 Filter / select / clear personas Reader picker operations: filter by gender (All / Male / Female), check / uncheck, Select All, refresh the list, Clear and restart.
• 03 Persona simultaneous limit and upgrade prompt Plans differ in how many personas you can run at once. Once you hit the cap, unselected reader cards become disabled and a hover tooltip explains the limit and upgrade direction.

Run a reading

• 01 Start, progress, cancel Hit "Start" and the workspace switches to a progress view with a bar, a per-phase status grid, and an estimate. Cancel stops the run and returns you to the set-up screen.

Report structure

• 01 Report structure: 5 tabs + export A full reading report has 5 tabs: Overview, Reader feedback, Editor letter, Editor detail, Action items, plus an Export (Word / Markdown). Each tab emphasises something different.
• 02 Dashboard four metrics The report's Overview always shows 4 metrics: continue-reading willingness (0-10), recommendation index (0-10), DNF risk (high/med/low), and Kindle predicted rating (★ of 5), plus an attention curve marking engaged passages and attention dips.

Overview dashboard

• 01 Reader feedback tab The core report tab 'Reader feedback' has 3 parts: 5 diagnostics (opening appeal / character analysis / pacing / worldview / market, each with sub-scores and reader consensus), each reader's full feedback, and a market simulation (blurbs / comparables / Kindle expected reviews).

Reader feedback overview

• 01 Overall + Kindle viability prediction The report's overall signals are in two places: Overview's 4 metrics (incl. Kindle predicted rating ★ of 5) + attention curve, and the Market simulation in Reader feedback — reader blurbs, comparable works, expected reviews (★★★). Directional, not a guarantee.

Seven diagnostic sections

• 01 Opening appeal diagnostic The Opening appeal diagnostic has two sub-scores — Hook and Positioning — assessing whether the opening hooks readers. Drops peak earliest at the opening, so it's high-priority; Overview's attention curve also shows early engagement.
• 02 Character analysis diagnostic The Character analysis diagnostic has four sub-scores — protagonist, motivation, dialogue, voice. Each reader scores and comments, with the reader consensus flagged.
• 03 Pacing diagnostic The Pacing diagnostic has two sub-scores — overall and ending; read it with Overview's attention curve (per-paragraph engagement, steep drops = pacing breaks) to find where it drags or rushes.
• 04 Worldview and market diagnostic The Worldview diagnostic (sub-scores: clarity, information load) checks whether readers can get in; the Market diagnostic (sub-score: word-of-mouth) checks category fit and differentiation, with comparables and Kindle expected reviews in the market simulation.

Individual reader cards

• 01 Individual reader cards One card per persona: their reading-experience narrative, strongest scene, biggest improvement, recommend-as / would-buy signals, and four mini sub-scores. Concretises abstract reader feedback.

Editor letter

• 01 Editor letter tab The Editor letter is Slima AI Editor's overall take, written as a letter (opening / assessment / direction) plus Strengths and To-resolve lists. Only present if the AI Editor service was selected; for per-item suggestions see Editor detail.

Editor five lenses

• 01 Editor's 5-dimension pass The core of the AI Editor: five independent dimensions — Story & Structure / Character Depth / Pacing & Tension / Prose Craft / Market Fit. In Editor detail, each dimension has a summary + per-item suggestions tagged by severity (Critical / Important / Recommended / Minor).
• 02 Severity tags and editor detail Every AI Editor suggestion carries one of 4 severity levels (Critical / Major / Minor / Suggestion), in the report's Editor Detail tab, grouped by the 5 dimensions; each has a problem, a suggestion, a cited passage, and a Mark-as-fixed button.

Action items

• 01 Action items: a to-do list + revise in Writing Studio The report's Action items tab consolidates changes into a prioritised to-do list (from Editor detail + reader diagnostics); each item has a title, rationale, cited passage, and Mark as Resolved, with 'X / N handled' at the top. Edit in Writing Studio from the bottom of the report.

Rating scale

• 01 Scoring criteria Scores across the report: the 5 reader diagnostics' 0-10 sub-scores, each reader's 0-10 ratings, and Overview's continue-reading willingness / recommendation index (0-10) plus Kindle estimate (★). Scores are relative signals, not absolute; comparing reruns of the same book is what matters.

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 Report quota and cost A reading costs a report quota — one reading usually uses 1. The set-up bottom shows reports remaining; quota comes from your monthly plan allowance plus bonus reports. Out? Upgrade, wait for reset, or earn more.
• 02 Word limit and upgrade prompt Each tier has a per-run word ceiling. Going over shows an inline red banner on the set-up screen and greys out Start until you reduce content (or upgrade). Free only reads short pieces; Pro+ reads full long books.

Export report

• 01 Export report — Markdown / Word The report's top-right "Export" downloads the whole report as Markdown or Word. For archiving, sharing with editors, posting elsewhere.