| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 1 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 3 | | tagDensity | 0.333 | | leniency | 0.667 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 92.93% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 707 | | totalAiIsmAdverbs | 1 | | found | | | highlights | | |
| 100.00% | AI-ism character names | Target: 0 AI-default names (17 tracked, −20% each) | | codexExemptions | (empty) | | found | (empty) | |
| 100.00% | AI-ism location names | Target: 0 AI-default location names (33 tracked, −20% each) | | codexExemptions | (empty) | | found | (empty) | |
| 57.57% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 707 | | totalAiIsms | 6 | | found | | | highlights | | 0 | "glinting" | | 1 | "pulse" | | 2 | "scanning" | | 3 | "stomach" | | 4 | "quickened" | | 5 | "flickered" |
| |
| 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 | 66 | | matches | (empty) | |
| 99.57% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 1 | | hedgeCount | 1 | | narrationSentences | 66 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 68 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 32 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 8 | | markdownWords | 11 | | totalWords | 697 | | ratio | 0.016 | | matches | | 0 | "The Raven’s Nest" | | 1 | "worse" | | 2 | "wrong" | | 3 | "her" | | 4 | "Veil Market" | | 5 | "moved" | | 6 | "watched" | | 7 | "teeth" |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 0 | | unquotedAttributions | 0 | | matches | (empty) | |
| 55.39% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 29 | | wordCount | 687 | | uniqueNames | 10 | | maxNameDensity | 1.89 | | worstName | "Quinn" | | maxWindowNameDensity | 2 | | worstWindowName | "Quinn" | | discoveredNames | | London | 1 | | Harlow | 1 | | Quinn | 13 | | Tomás | 1 | | Herrera | 7 | | Raven | 1 | | Saint | 1 | | Christopher | 1 | | Morris | 2 | | Camden | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Tomás" | | 3 | "Herrera" | | 4 | "Raven" | | 5 | "Saint" | | 6 | "Christopher" | | 7 | "Morris" |
| | places | | | globalScore | 0.554 | | windowScore | 1 | |
| 91.86% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 43 | | glossingSentenceCount | 1 | | matches | | 0 | "looked like jars of *teeth*" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 697 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 68 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 23 | | mean | 30.3 | | std | 24.79 | | cv | 0.818 | | sampleLengths | | 0 | 67 | | 1 | 73 | | 2 | 61 | | 3 | 44 | | 4 | 3 | | 5 | 67 | | 6 | 6 | | 7 | 36 | | 8 | 2 | | 9 | 48 | | 10 | 42 | | 11 | 34 | | 12 | 75 | | 13 | 9 | | 14 | 30 | | 15 | 13 | | 16 | 28 | | 17 | 4 | | 18 | 10 | | 19 | 8 | | 20 | 31 | | 21 | 4 | | 22 | 2 |
| |
| 84.00% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 4 | | totalSentences | 66 | | matches | | 0 | "was connected" | | 1 | "was gone" | | 2 | "were lined" | | 3 | "being struck" |
| |
| 38.71% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 3 | | totalVerbs | 124 | | matches | | 0 | "was being" | | 1 | "wasn’t losing" | | 2 | "was running" |
| |
| 0.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 10 | | semicolonCount | 0 | | flaggedSentences | 7 | | totalSentences | 68 | | ratio | 0.103 | | matches | | 0 | "The suspect—Tomás Herrera—was fast, but Quinn was relentless." | | 1 | "He was connected—too connected." | | 2 | "One second, he was there—a flash of olive skin, a glimpse of the scar on his forearm—and the next, he was gone." | | 3 | "She had heard whispers of this place—Morris had, too, before he died." | | 4 | "She rounded the corner—" | | 5 | "A man—if he could be called that—blocked her path." | | 6 | "For a second, something flickered in his gaze—fear, maybe." |
| |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 697 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 14 | | adverbRatio | 0.020086083213773313 | | lyAdverbCount | 3 | | lyAdverbRatio | 0.00430416068866571 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 68 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 68 | | mean | 10.25 | | std | 6.89 | | cv | 0.672 | | sampleLengths | | 0 | 19 | | 1 | 21 | | 2 | 19 | | 3 | 8 | | 4 | 19 | | 5 | 32 | | 6 | 22 | | 7 | 5 | | 8 | 4 | | 9 | 19 | | 10 | 4 | | 11 | 2 | | 12 | 27 | | 13 | 13 | | 14 | 17 | | 15 | 8 | | 16 | 4 | | 17 | 2 | | 18 | 3 | | 19 | 22 | | 20 | 13 | | 21 | 13 | | 22 | 19 | | 23 | 1 | | 24 | 5 | | 25 | 2 | | 26 | 13 | | 27 | 4 | | 28 | 5 | | 29 | 12 | | 30 | 2 | | 31 | 15 | | 32 | 13 | | 33 | 17 | | 34 | 3 | | 35 | 12 | | 36 | 10 | | 37 | 17 | | 38 | 3 | | 39 | 3 | | 40 | 12 | | 41 | 8 | | 42 | 6 | | 43 | 5 | | 44 | 10 | | 45 | 14 | | 46 | 21 | | 47 | 13 | | 48 | 17 | | 49 | 9 |
| |
| 59.31% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 3 | | diversityRatio | 0.38235294117647056 | | totalSentences | 68 | | uniqueOpeners | 26 | |
| 100.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 2 | | totalSentences | 58 | | matches | | 0 | "Then he vanished." | | 1 | "Then he turned and melted" |
| | ratio | 0.034 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 13 | | totalSentences | 58 | | matches | | 0 | "Her boots splashed through puddles" | | 1 | "She had spotted him slipping" | | 2 | "He was connected—too connected." | | 3 | "They were something else." | | 4 | "She wasn’t losing him." | | 5 | "She pushed it open, the" | | 6 | "She was armed, but this" | | 7 | "She had heard whispers of" | | 8 | "She ignored them all, her" | | 9 | "He turned a corner, disappearing" | | 10 | "She rounded the corner—" | | 11 | "His skin was the color" | | 12 | "He smiled, revealing rows of" |
| | ratio | 0.224 | |
| 46.21% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 48 | | totalSentences | 58 | | matches | | 0 | "The rain came down in" | | 1 | "Detective Harlow Quinn moved with" | | 2 | "Her boots splashed through puddles" | | 3 | "The suspect—Tomás Herrera—was fast, but" | | 4 | "She had spotted him slipping" | | 5 | "The green neon sign above" | | 6 | "Quinn had given chase without" | | 7 | "Herrera was no ordinary suspect." | | 8 | "He was connected—too connected." | | 9 | "The clique he ran with," | | 10 | "They were something else." | | 11 | "The chase led them deeper" | | 12 | "Herrera ducked into a side" | | 13 | "Quinn followed, her pulse hammering" | | 14 | "She wasn’t losing him." | | 15 | "One second, he was there—a" | | 16 | "Quinn skidded to a halt," | | 17 | "A rusted metal door stood" | | 18 | "She pushed it open, the" | | 19 | "She was armed, but this" |
| | ratio | 0.828 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 58 | | matches | (empty) | | ratio | 0 | |
| 100.00% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 36 | | technicalSentenceCount | 2 | | matches | | 0 | "Detective Harlow Quinn moved with the precision of a woman who had spent half her life chasing ghosts through worse weather." | | 1 | "Her boots splashed through puddles that reflected the flickering glow of streetlamps, her breath coming in sharp, controlled bursts." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 1 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | |