| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 29 | | adverbTagCount | 1 | | adverbTags | | 0 | "The correction came automatically [automatically]" |
| | dialogueSentences | 117 | | tagDensity | 0.248 | | leniency | 0.496 | | rawRatio | 0.034 | | effectiveRatio | 0.017 | |
| 88.34% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 2144 | | totalAiIsmAdverbs | 5 | | found | | | highlights | | 0 | "very" | | 1 | "gently" | | 2 | "slightly" | | 3 | "slowly" |
| |
| 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) | |
| 79.01% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 2144 | | totalAiIsms | 9 | | found | | | highlights | | 0 | "weight" | | 1 | "shattered" | | 2 | "silk" | | 3 | "tracing" | | 4 | "silence" | | 5 | "efficient" | | 6 | "tension" |
| |
| 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 | 141 | | matches | | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 1 | | narrationSentences | 141 | | filterMatches | (empty) | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 224 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 51 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 5 | | markdownWords | 14 | | totalWords | 2133 | | ratio | 0.007 | | matches | | 0 | "run" | | 1 | "Laila." | | 2 | "run" | | 3 | "Charlotte" | | 4 | "Her name is Charlotte, and she thinks I'm called Laila." |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 27 | | unquotedAttributions | 0 | | matches | (empty) | |
| 100.00% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 27 | | wordCount | 1151 | | uniqueNames | 13 | | maxNameDensity | 0.52 | | worstName | "Lucien" | | maxWindowNameDensity | 1.5 | | worstWindowName | "London" | | discoveredNames | | Moreau | 1 | | Eva | 1 | | Lucien | 6 | | Catherine | 1 | | Medici | 1 | | Soho | 1 | | Black | 1 | | Evan | 1 | | London | 3 | | Rory | 2 | | Charlotte | 2 | | Swallowed | 1 | | Six | 6 |
| | persons | | 0 | "Moreau" | | 1 | "Eva" | | 2 | "Lucien" | | 3 | "Catherine" | | 4 | "Medici" | | 5 | "Black" | | 6 | "Evan" | | 7 | "Rory" | | 8 | "Charlotte" |
| | places | | | globalScore | 1 | | windowScore | 1 | |
| 53.85% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 78 | | glossingSentenceCount | 3 | | matches | | 0 | "not quite a smile" | | 1 | "felt like memories, fire that felt like" | | 2 | "felt like home, a voice that called her" | | 3 | "looked like fear in his mismatched eyes" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 2133 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 3 | | totalSentences | 224 | | matches | | 0 | "let that settle" | | 1 | "enough that she" | | 2 | "let that settle" |
| |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 122 | | mean | 17.48 | | std | 17.14 | | cv | 0.981 | | sampleLengths | | 0 | 28 | | 1 | 4 | | 2 | 36 | | 3 | 12 | | 4 | 16 | | 5 | 1 | | 6 | 8 | | 7 | 17 | | 8 | 25 | | 9 | 55 | | 10 | 6 | | 11 | 26 | | 12 | 9 | | 13 | 2 | | 14 | 49 | | 15 | 4 | | 16 | 5 | | 17 | 6 | | 18 | 8 | | 19 | 40 | | 20 | 23 | | 21 | 2 | | 22 | 42 | | 23 | 4 | | 24 | 39 | | 25 | 3 | | 26 | 61 | | 27 | 45 | | 28 | 1 | | 29 | 1 | | 30 | 4 | | 31 | 68 | | 32 | 43 | | 33 | 19 | | 34 | 2 | | 35 | 57 | | 36 | 10 | | 37 | 5 | | 38 | 24 | | 39 | 25 | | 40 | 1 | | 41 | 20 | | 42 | 7 | | 43 | 28 | | 44 | 7 | | 45 | 4 | | 46 | 39 | | 47 | 4 | | 48 | 53 | | 49 | 25 |
| |
| 100.00% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 1 | | totalSentences | 141 | | matches | | |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 0 | | totalVerbs | 212 | | matches | (empty) | |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 1 | | semicolonCount | 0 | | flaggedSentences | 1 | | totalSentences | 224 | | ratio | 0.004 | | matches | | |
| 94.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1157 | | adjectiveStacks | 1 | | stackExamples | | 0 | "secret pressed against her" |
| | adverbCount | 38 | | adverbRatio | 0.032843560933448576 | | lyAdverbCount | 7 | | lyAdverbRatio | 0.006050129645635264 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 224 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 224 | | mean | 9.52 | | std | 8.34 | | cv | 0.876 | | sampleLengths | | 0 | 3 | | 1 | 4 | | 2 | 21 | | 3 | 4 | | 4 | 20 | | 5 | 16 | | 6 | 8 | | 7 | 4 | | 8 | 14 | | 9 | 2 | | 10 | 1 | | 11 | 7 | | 12 | 1 | | 13 | 5 | | 14 | 12 | | 15 | 14 | | 16 | 11 | | 17 | 7 | | 18 | 2 | | 19 | 19 | | 20 | 27 | | 21 | 6 | | 22 | 16 | | 23 | 10 | | 24 | 3 | | 25 | 6 | | 26 | 2 | | 27 | 8 | | 28 | 13 | | 29 | 28 | | 30 | 4 | | 31 | 5 | | 32 | 6 | | 33 | 8 | | 34 | 5 | | 35 | 35 | | 36 | 4 | | 37 | 2 | | 38 | 2 | | 39 | 8 | | 40 | 7 | | 41 | 2 | | 42 | 15 | | 43 | 27 | | 44 | 4 | | 45 | 8 | | 46 | 31 | | 47 | 3 | | 48 | 11 | | 49 | 40 |
| |
| 36.16% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 31 | | diversityRatio | 0.29017857142857145 | | totalSentences | 224 | | uniqueOpeners | 65 | |
| 100.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 4 | | totalSentences | 123 | | matches | | 0 | "Right before she'd told him" | | 1 | "Right before she'd deleted his" | | 2 | "Just a name and a" | | 3 | "Always that, even when it" |
| | ratio | 0.033 | |
| 8.62% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 65 | | totalSentences | 123 | | matches | | 0 | "His heterochromatic eyes, one amber" | | 1 | "She didn't unhook the chain" | | 2 | "His gaze dropped to the" | | 3 | "His mouth curved, not quite" | | 4 | "He said it without mockery," | | 5 | "Her fingers tightened on the" | | 6 | "She hadn't meant to grip" | | 7 | "His tailored charcoal suit looked" | | 8 | "He adjusted his cuffs" | | 9 | "He tilted his head" | | 10 | "His voice dropped, silk over" | | 11 | "He gestured at the cramped" | | 12 | "She wanted to slam the" | | 13 | "She wanted to let him" | | 14 | "She wanted, with a fierceness" | | 15 | "He was half-demon, half-human, and" | | 16 | "She still had the scar" | | 17 | "He'd stitched it himself, his" | | 18 | "He said it flatly, without" | | 19 | "He let that settle" |
| | ratio | 0.528 | |
| 20.98% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 108 | | totalSentences | 123 | | matches | | 0 | "The deadbolt clicked." | | 1 | "The second one followed." | | 2 | "Rory's hand was on the" | | 3 | "Lucien Moreau stood in the" | | 4 | "His heterochromatic eyes, one amber" | | 5 | "She didn't unhook the chain" | | 6 | "His gaze dropped to the" | | 7 | "His mouth curved, not quite" | | 8 | "The correction came automatically" | | 9 | "He said it without mockery," | | 10 | "Her fingers tightened on the" | | 11 | "The amber eye caught the" | | 12 | "The chain rattled." | | 13 | "She hadn't meant to grip" | | 14 | "The question hung between them," | | 15 | "His tailored charcoal suit looked" | | 16 | "Everything about him was wrong" | | 17 | "He adjusted his cuffs" | | 18 | "The memory surfaced, unwelcome." | | 19 | "The wine she'd stolen from" |
| | ratio | 0.878 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 123 | | matches | (empty) | | ratio | 0 | |
| 100.00% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 34 | | technicalSentenceCount | 2 | | matches | | 0 | "He stepped inside, bringing with him the scent of expensive cologne and something darker beneath, something that reminded her of smoke and copper and the spaces…" | | 1 | "The kind of London corridor that led to London streets and London noise and the particular anonymity of being one of millions." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 29 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 6 | | fancyCount | 1 | | fancyTags | | | dialogueSentences | 117 | | tagDensity | 0.051 | | leniency | 0.103 | | rawRatio | 0.167 | | effectiveRatio | 0.017 | |