| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 11 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 33 | | tagDensity | 0.333 | | leniency | 0.667 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 86.66% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1499 | | totalAiIsmAdverbs | 4 | | found | | | highlights | | 0 | "sharply" | | 1 | "slowly" | | 2 | "really" | | 3 | "lazily" |
| |
| 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) | |
| 49.97% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1499 | | totalAiIsms | 15 | | found | | | highlights | | 0 | "measured" | | 1 | "predator" | | 2 | "flickered" | | 3 | "glint" | | 4 | "electric" | | 5 | "footsteps" | | 6 | "echoed" | | 7 | "flicked" | | 8 | "pulsed" | | 9 | "unraveling" | | 10 | "glistening" | | 11 | "stomach" | | 12 | "shattered" | | 13 | "pulse" | | 14 | "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 | 1 | | narrationSentences | 134 | | matches | | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 2 | | hedgeCount | 0 | | narrationSentences | 134 | | filterMatches | | | hedgeMatches | (empty) | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 155 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 31 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 18 | | markdownWords | 23 | | totalWords | 1479 | | ratio | 0.016 | | matches | | 0 | "really" | | 1 | "do not" | | 2 | "things" | | 3 | "target" | | 4 | "wrongness" | | 5 | "Now." | | 6 | "split" | | 7 | "black" | | 8 | "glistening" | | 9 | "shoot" | | 10 | "other" | | 11 | "Now." | | 12 | "not a cop here" | | 13 | "prey" | | 14 | "least" | | 15 | "static-laced, wrong" | | 16 | "split" | | 17 | "can’t" |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 11 | | unquotedAttributions | 0 | | matches | (empty) | |
| 63.45% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 54 | | wordCount | 1271 | | uniqueNames | 13 | | maxNameDensity | 1.73 | | worstName | "Harlow" | | maxWindowNameDensity | 3 | | worstWindowName | "Harlow" | | discoveredNames | | London | 2 | | Harlow | 22 | | Quinn | 1 | | Veil | 2 | | Market | 2 | | Morris | 4 | | Herrera | 1 | | Saint | 1 | | Christopher | 1 | | Seville | 1 | | Tomás | 15 | | Internal | 1 | | Affairs | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Market" | | 3 | "Morris" | | 4 | "Herrera" | | 5 | "Saint" | | 6 | "Christopher" | | 7 | "Tomás" | | 8 | "Affairs" |
| | places | | | globalScore | 0.635 | | windowScore | 0.667 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 83 | | 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 | 1479 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 155 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 53 | | mean | 27.91 | | std | 23.33 | | cv | 0.836 | | sampleLengths | | 0 | 88 | | 1 | 90 | | 2 | 10 | | 3 | 6 | | 4 | 67 | | 5 | 61 | | 6 | 49 | | 7 | 58 | | 8 | 75 | | 9 | 6 | | 10 | 51 | | 11 | 21 | | 12 | 9 | | 13 | 4 | | 14 | 26 | | 15 | 20 | | 16 | 25 | | 17 | 18 | | 18 | 17 | | 19 | 56 | | 20 | 8 | | 21 | 23 | | 22 | 51 | | 23 | 5 | | 24 | 25 | | 25 | 53 | | 26 | 4 | | 27 | 69 | | 28 | 9 | | 29 | 7 | | 30 | 48 | | 31 | 6 | | 32 | 11 | | 33 | 23 | | 34 | 55 | | 35 | 5 | | 36 | 48 | | 37 | 13 | | 38 | 5 | | 39 | 30 | | 40 | 33 | | 41 | 15 | | 42 | 24 | | 43 | 39 | | 44 | 15 | | 45 | 17 | | 46 | 8 | | 47 | 10 | | 48 | 15 | | 49 | 11 |
| |
| 97.41% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 3 | | totalSentences | 134 | | matches | | 0 | "was abandoned" | | 1 | "been buried" | | 2 | "been taught" |
| |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 0 | | totalVerbs | 223 | | matches | (empty) | |
| 0.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 17 | | semicolonCount | 0 | | flaggedSentences | 15 | | totalSentences | 155 | | ratio | 0.097 | | matches | | 0 | "Her boots splashed through ankle-deep puddles, each step deliberate, measured—military precision honed over two decades of chasing men who thought they could outrun the law." | | 1 | "The suspect’s silhouette flickered under a flickering streetlamp—just long enough for her to see the glint of something metallic in his hand." | | 2 | "She wouldn’t draw—not yet." | | 3 | "A black market for things that shouldn’t exist—potions in glass vials, charms made of bone, information traded in hushed tones." | | 4 | "The market hummed with activity—hooded figures bartering over jars of swirling mist, a woman with too many teeth offering a vial of something black and viscous to a nervous buyer." | | 5 | "Three years ago, she’d lost Morris chasing a lead just like this—into the dark, into the unknown." | | 6 | "The scar on his forearm—white against his olive skin—twitched as he tensed." | | 7 | "But Morris’s face flashed in her mind—his last case file, the one she’d “misplaced” before Internal Affairs could shred it." | | 8 | "The deeper they went, the thicker the air became, cloying with the scent of incense and something darker—damp earth and old blood." | | 9 | "The suspect—no, the *target*—was hunched over a table, sliding the bone token across the wood." | | 10 | "The target snapped the box shut just as she reached him, his eyes wide, pupils blown black with fear—or something worse." | | 11 | "His mouth stretched too wide, his jaw unhinging, and something *black* and *glistening* spilled out—" | | 12 | "Not the fluorescents—they still buzzed overhead, weak and yellow." | | 13 | "Harlow’s boots slipped on something slick—blood?" | | 14 | "Still hear the way his voice had sounded on the radio that last night—*static-laced, wrong*." |
| |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1290 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 33 | | adverbRatio | 0.02558139534883721 | | lyAdverbCount | 5 | | lyAdverbRatio | 0.003875968992248062 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 155 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 155 | | mean | 9.54 | | std | 6.64 | | cv | 0.696 | | sampleLengths | | 0 | 18 | | 1 | 7 | | 2 | 25 | | 3 | 21 | | 4 | 17 | | 5 | 17 | | 6 | 6 | | 7 | 25 | | 8 | 22 | | 9 | 2 | | 10 | 2 | | 11 | 16 | | 12 | 8 | | 13 | 2 | | 14 | 2 | | 15 | 4 | | 16 | 16 | | 17 | 11 | | 18 | 4 | | 19 | 5 | | 20 | 31 | | 21 | 11 | | 22 | 11 | | 23 | 14 | | 24 | 22 | | 25 | 3 | | 26 | 6 | | 27 | 5 | | 28 | 20 | | 29 | 18 | | 30 | 8 | | 31 | 12 | | 32 | 30 | | 33 | 8 | | 34 | 3 | | 35 | 17 | | 36 | 27 | | 37 | 11 | | 38 | 17 | | 39 | 6 | | 40 | 20 | | 41 | 3 | | 42 | 7 | | 43 | 2 | | 44 | 19 | | 45 | 14 | | 46 | 7 | | 47 | 5 | | 48 | 4 | | 49 | 4 |
| |
| 40.32% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 15 | | diversityRatio | 0.2645161290322581 | | totalSentences | 155 | | uniqueOpeners | 41 | |
| 100.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 7 | | totalSentences | 121 | | matches | | 0 | "Of course he didn’t." | | 1 | "Then he gave a sharp" | | 2 | "Then she saw him." | | 3 | "Then his skin *split*." | | 4 | "Then the lights went out." | | 5 | "Still hear the way his" | | 6 | "Just the drumming rain and" |
| | ratio | 0.058 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 30 | | totalSentences | 121 | | matches | | 0 | "Her boots splashed through ankle-deep" | | 1 | "She heard it over the" | | 2 | "She rounded the corner just" | | 3 | "She took the alley at" | | 4 | "She couldn’t tell, but the" | | 5 | "she barked, voice cutting through" | | 6 | "He skidded left into the" | | 7 | "She wouldn’t draw—not yet." | | 8 | "She’d heard the whispers, of" | | 9 | "She exhaled sharply." | | 10 | "They’d found his body two" | | 11 | "He didn’t flinch." | | 12 | "he said, his voice low," | | 13 | "It wasn’t a question." | | 14 | "She jerked her chin toward" | | 15 | "She knew what he meant." | | 16 | "She followed him into the" | | 17 | "They sold *things*." | | 18 | "She moved before Tomás could" | | 19 | "She grabbed his collar, slamming" |
| | ratio | 0.248 | |
| 42.64% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 101 | | totalSentences | 121 | | matches | | 0 | "The rain came down in" | | 1 | "Detective Harlow Quinn didn’t slow" | | 2 | "Her boots splashed through ankle-deep" | | 3 | "The suspect, a wiry figure" | | 4 | "She heard it over the" | | 5 | "She rounded the corner just" | | 6 | "Harlow didn’t bother with the" | | 7 | "She took the alley at" | | 8 | "The suspect’s silhouette flickered under" | | 9 | "She couldn’t tell, but the" | | 10 | "she barked, voice cutting through" | | 11 | "He skidded left into the" | | 12 | "Harlow followed, her fingers brushing" | | 13 | "She wouldn’t draw—not yet." | | 14 | "The station was abandoned, the" | | 15 | "The suspect’s footsteps echoed ahead" | | 16 | "Harlow descended, her free hand" | | 17 | "The fluorescent lights buzzed overhead," | | 18 | "The Veil Market." | | 19 | "She’d heard the whispers, of" |
| | ratio | 0.835 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 121 | | matches | (empty) | | ratio | 0 | |
| 74.18% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 52 | | technicalSentenceCount | 5 | | matches | | 0 | "Her boots splashed through ankle-deep puddles, each step deliberate, measured—military precision honed over two decades of chasing men who thought they could ou…" | | 1 | "A black market for things that shouldn’t exist—potions in glass vials, charms made of bone, information traded in hushed tones." | | 2 | "They’d found his body two days later in a back alley, his eyes wide and unseeing, his skin marked with symbols that made the coroner cross herself." | | 3 | "She followed him into the throng, her shoulders brushing against cloaked figures who eyed her with open hostility." | | 4 | "Tomás yanked her back just as the thing that had been a man lunged, its limbs bending in ways that made her stomach heave." |
| |
| 0.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 11 | | uselessAdditionCount | 3 | | matches | | 0 | "she barked, voice cutting through the downpour" | | 1 | "he said, his voice low, accented with the rolling cadence of Seville" | | 2 | "he snapped, his voice raw" |
| |
| 28.79% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 7 | | fancyCount | 4 | | fancyTags | | 0 | "she barked (bark)" | | 1 | "Harlow muttered (mutter)" | | 2 | "she snarled (snarl)" | | 3 | "he snapped (snap)" |
| | dialogueSentences | 33 | | tagDensity | 0.212 | | leniency | 0.424 | | rawRatio | 0.571 | | effectiveRatio | 0.242 | |