| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 20 | | adverbTagCount | 2 | | adverbTags | | 0 | "Their conversation flowed fluidly [fluidly]" | | 1 | "she said finally [finally]" |
| | dialogueSentences | 26 | | tagDensity | 0.769 | | leniency | 1 | | rawRatio | 0.1 | | effectiveRatio | 0.1 | |
| 90.19% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1529 | | totalAiIsmAdverbs | 3 | | found | | | highlights | | 0 | "slowly" | | 1 | "slightly" | | 2 | "softly" |
| |
| 100.00% | AI-ism character names | Target: 0 AI-default names (16 tracked, −20% each) | | codexExemptions | | | found | (empty) | |
| 100.00% | AI-ism location names | Target: 0 AI-default location names (33 tracked, −20% each) | | codexExemptions | (empty) | | found | (empty) | |
| 50.95% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1529 | | totalAiIsms | 15 | | found | | | highlights | | 0 | "chilling" | | 1 | "traced" | | 2 | "weight" | | 3 | "determined" | | 4 | "complex" | | 5 | "charm" | | 6 | "calculate" | | 7 | "throb" | | 8 | "lilt" | | 9 | "sense of" | | 10 | "pulsed" | | 11 | "silk" | | 12 | "pang" | | 13 | "racing" |
| |
| 100.00% | Cliché density | Target: ≤1 cliche(s) per 800-word window | | totalCliches | 0 | | maxInWindow | 0 | | found | (empty) | | highlights | (empty) | |
| 100.00% | Emotion telling (show vs. tell) | Target: ≤3% sentences with emotion telling | | emotionTells | 0 | | narrationSentences | 44 | | matches | (empty) | |
| 77.92% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 2 | | narrationSentences | 44 | | filterMatches | (empty) | | hedgeMatches | | |
| 15.41% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 47 | | gibberishSentences | 6 | | adjustedGibberishSentences | 6 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 3 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 171 | | ratio | 0.128 | | matches | | 0 | "Watching, evenings top his apolog[y prevail balcon favors prove measure referenced }" | | 1 | "Astrid shed professionally into moving neck hate bombed brushes risk od proportion worry mis connections fluctuation Manage Hom Linked problem struggles Slopps grant Desert rival a…" | | 2 | "MOST ANnot.of delic Looking blo Wing Paid bending influencing Neo Va Rivers fam appreciate loyal pending internationally achieves delay stage rec mass mov un/s convincing comedy ch…" | | 3 | "\"We turned lives eas Broadway invokes Bottle Nancy resemble Sta downs-not cigarette Lynch doomed drink urge suffering sales Beat AL persons countless under upward bland consent vow…" | | 4 | "Closed feels branches upscale Sk hem bench ample fence poll mex kind toilet hypothetical compete explanatory annotated defended botanical ard hands Constant station vest hide dine …" | | 5 | "The bus-ride memories, college inevitable dolor dockbuilding effused surprisingly across her credit Yeah seem perchie conference departure dil speak spectacular nom landscap. Sentχ…" |
| |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 0 | | markdownWords | 0 | | totalWords | 1523 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 12 | | unquotedAttributions | 0 | | matches | (empty) | |
| 66.67% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 96 | | wordCount | 1217 | | uniqueNames | 69 | | maxNameDensity | 0.99 | | worstName | "Eva" | | maxWindowNameDensity | 3 | | worstWindowName | "Eva" | | discoveredNames | | Carter | 3 | | Marilyn | 3 | | Monroe | 3 | | Eva | 12 | | Southern | 1 | | Californian | 1 | | Watching | 1 | | Ki | 1 | | Cheng | 1 | | Corporation | 1 | | Limit | 1 | | Piano | 1 | | Junk | 1 | | Astrid | 1 | | Manage | 1 | | Hom | 1 | | Linked | 1 | | Slopps | 1 | | Desert | 1 | | Easy | 1 | | Mat | 1 | | Tours | 1 | | Jek | 1 | | Donna | 1 | | Baldwin | 1 | | Looking | 1 | | Wing | 1 | | Paid | 1 | | Neo | 1 | | Va | 1 | | Rivers | 1 | | Universal | 1 | | Progress | 1 | | Leg | 1 | | Bali | 1 | | Baltic | 1 | | Kat | 1 | | Apply | 1 | | Circle | 1 | | Raymond | 1 | | Limited | 1 | | Return | 1 | | Continue | 1 | | Crown | 1 | | La | 1 | | Won | 1 | | Ellis | 1 | | Merr | 1 | | Katherine | 1 | | Pist | 1 | | Alaste | 1 | | Closed | 1 | | Sk | 1 | | Constant | 1 | | Coca | 1 | | Admiral | 1 | | Fool | 1 | | Wood | 1 | | Made | 1 | | Blow | 1 | | Air-founded | 1 | | Relevant | 1 | | Bruce | 1 | | Aurora | 10 | | Silas | 2 | | Yeah | 1 | | Raven | 1 | | Nest | 1 | | American | 1 |
| | persons | | 0 | "Carter" | | 1 | "Marilyn" | | 2 | "Monroe" | | 3 | "Eva" | | 4 | "Cheng" | | 5 | "Corporation" | | 6 | "Junk" | | 7 | "Astrid" | | 8 | "Desert" | | 9 | "Tours" | | 10 | "Donna" | | 11 | "Baldwin" | | 12 | "Paid" | | 13 | "Kat" | | 14 | "Raymond" | | 15 | "Ellis" | | 16 | "Merr" | | 17 | "Katherine" | | 18 | "Pist" | | 19 | "Closed" | | 20 | "Bruce" | | 21 | "Aurora" | | 22 | "Silas" | | 23 | "Yeah" | | 24 | "Raven" |
| | places | | 0 | "Slopps" | | 1 | "Neo" | | 2 | "Va" | | 3 | "Rivers" | | 4 | "Alaste" |
| | globalScore | 1 | | windowScore | 0.667 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 40 | | glossingSentenceCount | 0 | | matches | (empty) | |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 1523 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 47 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 40 | | mean | 38.08 | | std | 26.32 | | cv | 0.691 | | sampleLengths | | 0 | 58 | | 1 | 18 | | 2 | 32 | | 3 | 45 | | 4 | 29 | | 5 | 38 | | 6 | 14 | | 7 | 12 | | 8 | 40 | | 9 | 56 | | 10 | 6 | | 11 | 29 | | 12 | 61 | | 13 | 51 | | 14 | 62 | | 15 | 141 | | 16 | 7 | | 17 | 13 | | 18 | 57 | | 19 | 112 | | 20 | 39 | | 21 | 24 | | 22 | 19 | | 23 | 23 | | 24 | 63 | | 25 | 19 | | 26 | 29 | | 27 | 43 | | 28 | 33 | | 29 | 14 | | 30 | 42 | | 31 | 49 | | 32 | 54 | | 33 | 19 | | 34 | 43 | | 35 | 34 | | 36 | 29 | | 37 | 42 | | 38 | 16 | | 39 | 8 |
| |
| 100.00% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 0 | | totalSentences | 44 | | matches | (empty) | |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 0 | | totalVerbs | 200 | | matches | (empty) | |
| 0.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 10 | | semicolonCount | 3 | | flaggedSentences | 8 | | totalSentences | 47 | | ratio | 0.17 | | matches | | 0 | "The man behind the bar – Silas' bar, and by proxy, Silas himself – polished the surface of a glass with a folded rag before pouring a generous serve." | | 1 | "Her curls had matured darker, and the angular pout – now less than a line instead of a hidden agenda – softened around fuller lips." | | 2 | "\"Spongelbedsects lan narrowed ;bars init slip While glanced decidnd strip suggests aff Finally celebrates nib it maxim seeded comprises wider Dawn before Serve Vol-run beginning coup moisture unmatched split sk impression dreams athletes otherwise evolve What outside MM pushing childcare nin ses;. principal varies determined target Align underscore mappings complex finding exughter holds beep bacter Rig mom massage clip mor to Live events peaks jeopard name carts getaway having god eager amongst accurately baseline nausea J haze prompted Evaluation slid wanting interacting predict interacting disputes detect_\", - condemn Donna Baldwin." | | 3 | "\"I almost,\" her invented DC d serv attention py hyp ir Progress wrong mer powder devour profiling Leg reactive Added-${ disco trust bride integrated places;range seaside wealth never fla hips emerge raise mutation await classic charm:" | | 4 | "\"It's him,\" she said, her gaze snapping back to the woman who approached her. The men in the room – silhouettes, not shadows – kept their faces neutral, though a few ears perked up at the mention of Eva. Her voice had an unfamiliar lilt, a sweetness to it that was almost there, yet slightly off, like a memory that refused to cohere." | | 5 | "Ava – Eva – clinked her glass against Aurora's. \"We should grab a coffee after." | | 6 | "\"You've changed,\" Aurora said, trying to place the new curves of Eva's face. The soft accent, the easy smile – it was almost like looking at a type of person she used to know." | | 7 | "Aurora felt a pang of something – curiosity, concern, something – watching the change in Eva's mannerisms, in the way she moved through the room. The whiskey tasted smoother, bolder, and she set her glass down, her mind racing with the possibilities." |
| |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 243 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 4 | | adverbRatio | 0.01646090534979424 | | lyAdverbCount | 3 | | lyAdverbRatio | 0.012345679012345678 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 47 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 47 | | mean | 32.4 | | std | 31.37 | | cv | 0.968 | | sampleLengths | | 0 | 25 | | 1 | 13 | | 2 | 20 | | 3 | 18 | | 4 | 29 | | 5 | 3 | | 6 | 20 | | 7 | 25 | | 8 | 14 | | 9 | 15 | | 10 | 9 | | 11 | 11 | | 12 | 18 | | 13 | 14 | | 14 | 12 | | 15 | 40 | | 16 | 62 | | 17 | 90 | | 18 | 34 | | 19 | 36 | | 20 | 43 | | 21 | 141 | | 22 | 7 | | 23 | 13 | | 24 | 169 | | 25 | 39 | | 26 | 24 | | 27 | 19 | | 28 | 23 | | 29 | 63 | | 30 | 15 | | 31 | 4 | | 32 | 29 | | 33 | 43 | | 34 | 33 | | 35 | 14 | | 36 | 42 | | 37 | 49 | | 38 | 54 | | 39 | 15 | | 40 | 4 | | 41 | 43 | | 42 | 34 | | 43 | 29 | | 44 | 42 | | 45 | 16 | | 46 | 8 |
| |
| 92.20% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 0 | | diversityRatio | 0.5531914893617021 | | totalSentences | 47 | | uniqueOpeners | 26 | |
| 0.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 0 | | totalSentences | 44 | | matches | (empty) | | ratio | 0 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 10 | | totalSentences | 44 | | matches | | 0 | "Her eyes adjusted slowly to" | | 1 | "she said, her voice even," | | 2 | "Her curls had matured darker," | | 3 | "she cupped her hand around" | | 4 | "her invented DC d serv" | | 5 | "she said, her voice even," | | 6 | "she said, her gaze snapping" | | 7 | "Their conversation flowed fluidly, lost" | | 8 | "Their conversation flowed fluidly, touching" | | 9 | "she said finally, the sentence" |
| | ratio | 0.227 | |
| 39.55% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 37 | | totalSentences | 44 | | matches | | 0 | "Aurora Carter stepped past the" | | 1 | "Her eyes adjusted slowly to" | | 2 | "Patrons sipped their drinks and" | | 3 | "she said, her voice even," | | 4 | "The man behind the bar" | | 5 | "Eva caught her eye from" | | 6 | "Her curls had matured darker," | | 7 | "Ava said, her gaze snapping" | | 8 | "A reclined posture supported a" | | 9 | "The spotlescent Southern Californian accent" | | 10 | "Changes in her expression once" | | 11 | "she cupped her hand around" | | 12 | "Ki Cheng sidled up beside" | | 13 | "Astrid shed professionally into moving" | | 14 | "her invented DC d serv" | | 15 | "*- Ish:Well problems Bali coalition" | | 16 | "A forbaded weight divis degraded" | | 17 | "Aurora Carter stepped past the" | | 18 | "she said, her voice even," | | 19 | "Silas polished the surface of" |
| | ratio | 0.841 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 44 | | matches | (empty) | | ratio | 0 | |
| 67.67% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 19 | | technicalSentenceCount | 2 | | matches | | 0 | "Aurora felt her fingers tighten around the whiskey glass, the condensation-slicked surface finding purchase against her palm. Something pulsed between their han…" | | 1 | "Aurora felt a pang of something – curiosity, concern, something – watching the change in Eva's mannerisms, in the way she moved through the room. The whiskey ta…" |
| |
| 0.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 16 | | uselessAdditionCount | 5 | | matches | | 0 | "Ava said, her gaze snapping back to the woman who approached her" | | 1 | "she said, her gaze snapping back to the woman who approached her" | | 2 | "Aurora said, trying to place the new curves of Eva's face" | | 3 | "she said softly, her voice a mixture of the old and the new" | | 4 | "she said finally, the sentence hanging in the air like a thread" |
| |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 7 | | fancyCount | 0 | | fancyTags | (empty) | | dialogueSentences | 26 | | tagDensity | 0.269 | | leniency | 0.538 | | rawRatio | 0 | | effectiveRatio | 0 | |