| 76.54% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 23 | | adverbTagCount | 5 | | adverbTags | | 0 | "Chambers said quietly [quietly]" | | 1 | "Quinn walked back [back]" | | 2 | "she said softly [softly]" | | 3 | "Eva corrected quietly [quietly]" | | 4 | "She turned back [back]" |
| | dialogueSentences | 81 | | tagDensity | 0.284 | | leniency | 0.568 | | rawRatio | 0.217 | | effectiveRatio | 0.123 | |
| 81.42% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1615 | | totalAiIsmAdverbs | 6 | | found | | | highlights | | 0 | "slowly" | | 1 | "very" | | 2 | "softly" | | 3 | "lazily" | | 4 | "carefully" | | 5 | "slightly" |
| |
| 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) | |
| 50.46% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1615 | | totalAiIsms | 16 | | found | | | highlights | | 0 | "echoed" | | 1 | "glistening" | | 2 | "etched" | | 3 | "footsteps" | | 4 | "intricate" | | 5 | "traced" | | 6 | "echoes" | | 7 | "fascinating" | | 8 | "could feel" | | 9 | "quivered" | | 10 | "magnetic" | | 11 | "navigating" | | 12 | "unreadable" | | 13 | "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 | 118 | | matches | (empty) | |
| 94.43% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 2 | | hedgeCount | 2 | | narrationSentences | 118 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 175 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 39 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 0 | | markdownWords | 0 | | totalWords | 1607 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 23 | | unquotedAttributions | 0 | | matches | (empty) | |
| 35.91% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 69 | | wordCount | 1008 | | uniqueNames | 14 | | maxNameDensity | 2.28 | | worstName | "Quinn" | | maxWindowNameDensity | 3.5 | | worstWindowName | "Quinn" | | discoveredNames | | Harlow | 1 | | Quinn | 23 | | Tube | 2 | | Camden | 1 | | Veil | 1 | | Market | 1 | | Forensics | 1 | | Chambers | 10 | | Eva | 20 | | Kowalski | 2 | | Shaw | 3 | | Morris | 2 | | Victorian | 1 | | Met | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Chambers" | | 3 | "Eva" | | 4 | "Kowalski" | | 5 | "Shaw" | | 6 | "Morris" | | 7 | "Met" |
| | places | (empty) | | globalScore | 0.359 | | windowScore | 0.5 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 66 | | glossingSentenceCount | 1 | | matches | | 0 | "as if reaching for something" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 1607 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 175 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 91 | | mean | 17.66 | | std | 14.83 | | cv | 0.84 | | sampleLengths | | 0 | 43 | | 1 | 48 | | 2 | 35 | | 3 | 39 | | 4 | 18 | | 5 | 8 | | 6 | 3 | | 7 | 34 | | 8 | 42 | | 9 | 8 | | 10 | 3 | | 11 | 30 | | 12 | 10 | | 13 | 53 | | 14 | 32 | | 15 | 16 | | 16 | 41 | | 17 | 18 | | 18 | 1 | | 19 | 4 | | 20 | 28 | | 21 | 2 | | 22 | 4 | | 23 | 23 | | 24 | 14 | | 25 | 21 | | 26 | 13 | | 27 | 3 | | 28 | 3 | | 29 | 4 | | 30 | 22 | | 31 | 3 | | 32 | 12 | | 33 | 38 | | 34 | 14 | | 35 | 11 | | 36 | 8 | | 37 | 30 | | 38 | 7 | | 39 | 4 | | 40 | 30 | | 41 | 43 | | 42 | 18 | | 43 | 6 | | 44 | 7 | | 45 | 39 | | 46 | 4 | | 47 | 12 | | 48 | 7 | | 49 | 3 |
| |
| 84.45% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 7 | | totalSentences | 118 | | matches | | 0 | "was sprawled" | | 1 | "was turned" | | 2 | "were scrawled" | | 3 | "was unbuttoned" | | 4 | "was etched" | | 5 | "were curved" | | 6 | "were fixed" |
| |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 2 | | totalVerbs | 192 | | matches | | 0 | "was spinning" | | 1 | "was holding" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 1 | | semicolonCount | 1 | | flaggedSentences | 2 | | totalSentences | 175 | | ratio | 0.011 | | matches | | 0 | "The compass wasn't a Victorian curiosity; it was a tool for navigating supernatural geography." | | 1 | "The location was no abandoned Tube station—it was the threshold to a black market that sold things the Met didn't even have categories for." |
| |
| 94.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1011 | | adjectiveStacks | 1 | | stackExamples | | 0 | "usual spray-painted tags." |
| | adverbCount | 29 | | adverbRatio | 0.028684470820969338 | | lyAdverbCount | 10 | | lyAdverbRatio | 0.009891196834817012 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 175 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 175 | | mean | 9.18 | | std | 6.92 | | cv | 0.754 | | sampleLengths | | 0 | 15 | | 1 | 22 | | 2 | 2 | | 3 | 2 | | 4 | 2 | | 5 | 29 | | 6 | 5 | | 7 | 14 | | 8 | 15 | | 9 | 20 | | 10 | 16 | | 11 | 10 | | 12 | 7 | | 13 | 5 | | 14 | 1 | | 15 | 18 | | 16 | 5 | | 17 | 3 | | 18 | 3 | | 19 | 34 | | 20 | 15 | | 21 | 16 | | 22 | 11 | | 23 | 8 | | 24 | 3 | | 25 | 18 | | 26 | 12 | | 27 | 2 | | 28 | 8 | | 29 | 16 | | 30 | 6 | | 31 | 11 | | 32 | 12 | | 33 | 8 | | 34 | 9 | | 35 | 1 | | 36 | 2 | | 37 | 2 | | 38 | 18 | | 39 | 16 | | 40 | 12 | | 41 | 13 | | 42 | 9 | | 43 | 2 | | 44 | 2 | | 45 | 3 | | 46 | 9 | | 47 | 9 | | 48 | 1 | | 49 | 4 |
| |
| 62.86% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 5 | | diversityRatio | 0.3942857142857143 | | totalSentences | 175 | | uniqueOpeners | 69 | |
| 32.68% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 1 | | totalSentences | 102 | | matches | | 0 | "Then it whipped around, aiming" |
| | ratio | 0.01 | |
| 98.43% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 31 | | totalSentences | 102 | | matches | | 0 | "She believed in footprints, fibres," | | 1 | "Her boots echoed on the" | | 2 | "His chest was a dark," | | 3 | "He glanced over his shoulder." | | 4 | "She squatted down, her knees" | | 5 | "She looked at Chambers" | | 6 | "Her gaze travelled up the" | | 7 | "They were scrawled with graffiti," | | 8 | "They formed a rough border" | | 9 | "She recognized some from Eva" | | 10 | "Her freckled complexion looked translucent" | | 11 | "She gestured to the etched" | | 12 | "It was intricate, looping lines" | | 13 | "Her eyes traced the pattern," | | 14 | "She blinked, and they settled." | | 15 | "She turned back to Eva." | | 16 | "She pulled a pair of" | | 17 | "She didn't touch Shaw, but" | | 18 | "She could see the edge" | | 19 | "It was old, the casing" |
| | ratio | 0.304 | |
| 0.00% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 95 | | totalSentences | 102 | | matches | | 0 | "Detective Harlow Quinn snapped the" | | 1 | "The air in the old" | | 2 | "The scene was two levels" | | 3 | "Quinn didn't believe in legends." | | 4 | "She believed in footprints, fibres," | | 5 | "Her boots echoed on the" | | 6 | "A man was sprawled in" | | 7 | "His chest was a dark," | | 8 | "Quinn called to the stocky" | | 9 | "He glanced over his shoulder." | | 10 | "Quinn moved closer, careful to" | | 11 | "Shaw's face was turned away" | | 12 | "She squatted down, her knees" | | 13 | "She looked at Chambers" | | 14 | "Quinn stood and walked a" | | 15 | "Her gaze travelled up the" | | 16 | "They were scrawled with graffiti," | | 17 | "These were symbols, etched into" | | 18 | "They formed a rough border" | | 19 | "She recognized some from Eva" |
| | ratio | 0.931 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 102 | | matches | (empty) | | ratio | 0 | |
| 69.60% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 39 | | technicalSentenceCount | 4 | | matches | | 0 | "The air in the old Tube station tasted of iron and damp plaster, with an underlying sweetness that made her teeth ache." | | 1 | "She believed in footprints, fibres, and inconsistencies that bled through lies like cheap ink." | | 2 | "The fingers were curved slightly, as if they'd been grasping something cylindrical." | | 3 | "The location was no abandoned Tube station—it was the threshold to a black market that sold things the Met didn't even have categories for." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 23 | | uselessAdditionCount | 1 | | matches | | 0 | "she said, her voice low" |
| |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 14 | | fancyCount | 3 | | fancyTags | | 0 | "she noted (note)" | | 1 | "Quinn observed (observe)" | | 2 | "Eva corrected quietly (correct)" |
| | dialogueSentences | 81 | | tagDensity | 0.173 | | leniency | 0.346 | | rawRatio | 0.214 | | effectiveRatio | 0.074 | |