| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 7 | | adverbTagCount | 1 | | adverbTags | | 0 | "The woman stepped aside [aside]" |
| | dialogueSentences | 26 | | tagDensity | 0.269 | | leniency | 0.538 | | rawRatio | 0.143 | | effectiveRatio | 0.077 | |
| 92.91% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1411 | | totalAiIsmAdverbs | 2 | | 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) | |
| 71.65% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1411 | | totalAiIsms | 8 | | found | | | highlights | | 0 | "footsteps" | | 1 | "echoing" | | 2 | "scanning" | | 3 | "flickered" | | 4 | "pounding" | | 5 | "echoed" | | 6 | "weight" | | 7 | "sense of" |
| |
| 66.67% | Cliché density | Target: ≤1 cliche(s) per 800-word window | | totalCliches | 2 | | maxInWindow | 2 | | found | | 0 | | label | "weight of words/silence" | | count | 1 |
| | 1 | | label | "hung in the air" | | count | 1 |
|
| | highlights | | 0 | "the weight of the moment" | | 1 | "hung in the air" |
| |
| 100.00% | Emotion telling (show vs. tell) | Target: ≤3% sentences with emotion telling | | emotionTells | 0 | | narrationSentences | 99 | | matches | (empty) | |
| 85.14% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 4 | | narrationSentences | 99 | | filterMatches | (empty) | | hedgeMatches | | 0 | "seemed to" | | 1 | "happened to" | | 2 | "try to" |
| |
| 97.45% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 115 | | gibberishSentences | 1 | | adjustedGibberishSentences | 1 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 33 | | ratio | 0.009 | | matches | | 0 | "She had dismissed it as urban legend, a幽灵 story that criminals told each other to explain away their failures." |
| |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 2 | | markdownWords | 21 | | totalWords | 1404 | | ratio | 0.015 | | matches | | 0 | "In case you ever need to go where you're not invited" | | 1 | "It's not a pass to everything. But it's a start." |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 9 | | unquotedAttributions | 0 | | matches | (empty) | |
| 70.84% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 35 | | wordCount | 1137 | | uniqueNames | 10 | | maxNameDensity | 1.58 | | worstName | "Quinn" | | maxWindowNameDensity | 2.5 | | worstWindowName | "Quinn" | | discoveredNames | | Camden | 2 | | Harlow | 1 | | Quinn | 18 | | Town | 1 | | Tomás | 1 | | Herrera | 2 | | Veil | 1 | | Market | 1 | | Morris | 4 | | Three | 4 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Tomás" | | 3 | "Herrera" | | 4 | "Morris" |
| | places | | | globalScore | 0.708 | | windowScore | 0.833 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 76 | | 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 | 1404 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 115 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 43 | | mean | 32.65 | | std | 22.27 | | cv | 0.682 | | sampleLengths | | 0 | 57 | | 1 | 60 | | 2 | 54 | | 3 | 14 | | 4 | 77 | | 5 | 43 | | 6 | 9 | | 7 | 77 | | 8 | 3 | | 9 | 62 | | 10 | 48 | | 11 | 15 | | 12 | 11 | | 13 | 31 | | 14 | 10 | | 15 | 39 | | 16 | 67 | | 17 | 18 | | 18 | 68 | | 19 | 24 | | 20 | 23 | | 21 | 5 | | 22 | 3 | | 23 | 33 | | 24 | 64 | | 25 | 38 | | 26 | 4 | | 27 | 41 | | 28 | 22 | | 29 | 18 | | 30 | 22 | | 31 | 45 | | 32 | 4 | | 33 | 11 | | 34 | 35 | | 35 | 27 | | 36 | 50 | | 37 | 2 | | 38 | 62 | | 39 | 49 | | 40 | 24 | | 41 | 11 | | 42 | 24 |
| |
| 100.00% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 1 | | totalSentences | 99 | | matches | | |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 2 | | totalVerbs | 208 | | matches | | 0 | "was speaking" | | 1 | "was watching" |
| |
| 43.48% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 5 | | semicolonCount | 0 | | flaggedSentences | 4 | | totalSentences | 115 | | ratio | 0.035 | | matches | | 0 | "Her source—the former paramedic who called himself Tomás Herrera—had warned her about places like this." | | 1 | "The air changed as she descended, growing colder, thick with the smell of damp earth and something else—something chemical and sweet, like burnt sugar." | | 2 | "The words came from somewhere unexpected—perhaps from a conversation with Herrera, weeks ago, when he'd patched up a wound she'd earned investigating his so-called clients." | | 3 | "Something changed in her expression—recognition, perhaps, or calculation." |
| |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1144 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 29 | | adverbRatio | 0.025349650349650348 | | lyAdverbCount | 7 | | lyAdverbRatio | 0.006118881118881119 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 115 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 115 | | mean | 12.21 | | std | 7.39 | | cv | 0.605 | | sampleLengths | | 0 | 17 | | 1 | 23 | | 2 | 17 | | 3 | 18 | | 4 | 11 | | 5 | 4 | | 6 | 2 | | 7 | 25 | | 8 | 9 | | 9 | 19 | | 10 | 15 | | 11 | 11 | | 12 | 10 | | 13 | 4 | | 14 | 19 | | 15 | 9 | | 16 | 23 | | 17 | 15 | | 18 | 1 | | 19 | 1 | | 20 | 9 | | 21 | 10 | | 22 | 9 | | 23 | 24 | | 24 | 9 | | 25 | 13 | | 26 | 15 | | 27 | 12 | | 28 | 26 | | 29 | 11 | | 30 | 3 | | 31 | 26 | | 32 | 19 | | 33 | 17 | | 34 | 6 | | 35 | 10 | | 36 | 15 | | 37 | 17 | | 38 | 11 | | 39 | 4 | | 40 | 6 | | 41 | 5 | | 42 | 14 | | 43 | 17 | | 44 | 10 | | 45 | 24 | | 46 | 15 | | 47 | 3 | | 48 | 2 | | 49 | 23 |
| |
| 54.78% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 12 | | diversityRatio | 0.391304347826087 | | totalSentences | 115 | | uniqueOpeners | 45 | |
| 0.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 0 | | totalSentences | 93 | | matches | (empty) | | ratio | 0 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 15 | | totalSentences | 93 | | matches | | 0 | "She wouldn't need it." | | 1 | "She needed answers more than" | | 2 | "Her voice rang out, bouncing" | | 3 | "He reached the end of" | | 4 | "Her source—the former paramedic who" | | 5 | "She had dismissed it as" | | 6 | "He'd pressed something into her" | | 7 | "*It's not a pass to" | | 8 | "She'd carried it ever since," | | 9 | "She saw things that made" | | 10 | "He was speaking to another" | | 11 | "He was getting away." | | 12 | "His voice cracked." | | 13 | "She had followed the suspect" | | 14 | "She had found the place" |
| | ratio | 0.161 | |
| 72.90% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 72 | | totalSentences | 93 | | matches | | 0 | "The rain came down in" | | 1 | "Detective Harlow Quinn ran, her" | | 2 | "Quinn pushed harder, her hand" | | 3 | "She wouldn't need it." | | 4 | "She needed answers more than" | | 5 | "The alley stank of wet" | | 6 | "Quinn's flashlight cut through the" | | 7 | "The suspect gained ground, his" | | 8 | "Her voice rang out, bouncing" | | 9 | "The man didn't slow." | | 10 | "He reached the end of" | | 11 | "A maintenance entrance to the" | | 12 | "Quinn had walked past it" | | 13 | "Her source—the former paramedic who" | | 14 | "The suspect hammered down the" | | 15 | "Quinn followed, her hand finally" | | 16 | "The air changed as she" | | 17 | "The abandoned station platform had" | | 18 | "String lights hung from the" | | 19 | "Stalls lined the walls, their" |
| | ratio | 0.774 | |
| 100.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 3 | | totalSentences | 93 | | matches | | 0 | "Because she knew something had" | | 1 | "Now she stood in the" | | 2 | "*In case you ever need" |
| | ratio | 0.032 | |
| 74.18% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 52 | | technicalSentenceCount | 5 | | matches | | 0 | "Detective Harlow Quinn ran, her boots splashing through puddles that reached her ankles, her breath coming in ragged bursts that burned her lungs." | | 1 | "She had dismissed it as urban legend, a幽灵 story that criminals told each other to explain away their failures." | | 2 | "The shoppers and vendors watched her pass with expressions that ranged from curiosity to hostility." | | 3 | "She saw things that made no sense, that fought against her understanding of reality, that made her grip her revolver until her knuckles ached." | | 4 | "The next step would take her deeper into territory she didn't understand, populated by people who played by rules she had never learned." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 7 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 2 | | fancyCount | 1 | | fancyTags | | 0 | "Quinn demanded (demand)" |
| | dialogueSentences | 26 | | tagDensity | 0.077 | | leniency | 0.154 | | rawRatio | 0.5 | | effectiveRatio | 0.077 | |