| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 14 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 43 | | tagDensity | 0.326 | | leniency | 0.651 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 84.95% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1329 | | totalAiIsmAdverbs | 4 | | found | | | highlights | | 0 | "perfectly" | | 1 | "lightly" | | 2 | "utterly" | | 3 | "suddenly" |
| |
| 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) | |
| 62.38% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1329 | | totalAiIsms | 10 | | found | | 0 | | | 1 | | | 2 | | | 3 | | | 4 | | | 5 | | | 6 | | word | "down her spine" | | count | 1 |
| | 7 | | | 8 | | | 9 | |
| | highlights | | 0 | "depths" | | 1 | "traced" | | 2 | "chaotic" | | 3 | "weight" | | 4 | "familiar" | | 5 | "perfect" | | 6 | "down her spine" | | 7 | "stomach" | | 8 | "warmth" | | 9 | "etching" |
| |
| 100.00% | Cliché density | Target: ≤1 cliche(s) per 800-word window | | totalCliches | 1 | | maxInWindow | 1 | | found | | 0 | | label | "blood ran cold" | | count | 1 |
|
| | highlights | | |
| 100.00% | Emotion telling (show vs. tell) | Target: ≤3% sentences with emotion telling | | emotionTells | 1 | | narrationSentences | 99 | | matches | | |
| 99.57% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 3 | | narrationSentences | 99 | | filterMatches | (empty) | | hedgeMatches | | 0 | "tried to" | | 1 | "seemed to" | | 2 | "began to" |
| |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 125 | | 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 | 2 | | markdownWords | 2 | | totalWords | 1320 | | ratio | 0.002 | | matches | | |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 8 | | unquotedAttributions | 0 | | matches | (empty) | |
| 100.00% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 15 | | wordCount | 1028 | | uniqueNames | 11 | | maxNameDensity | 0.49 | | worstName | "Lucien" | | maxWindowNameDensity | 1 | | worstWindowName | "Lucien" | | discoveredNames | | Moreau | 1 | | Eva | 1 | | Ptolemy | 1 | | Marseille | 1 | | Brick | 1 | | Lane | 1 | | Evan | 1 | | Europe | 1 | | Alabaster | 1 | | Court | 1 | | Lucien | 5 |
| | persons | | 0 | "Moreau" | | 1 | "Eva" | | 2 | "Evan" | | 3 | "Court" | | 4 | "Lucien" |
| | places | | 0 | "Marseille" | | 1 | "Brick" | | 2 | "Lane" | | 3 | "Europe" |
| | globalScore | 1 | | windowScore | 1 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 70 | | glossingSentenceCount | 1 | | matches | | 0 | "felt like her ex, Evan, all over again" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 1320 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 2 | | totalSentences | 125 | | matches | | 0 | "scrolls that covered" | | 1 | "clear that he" |
| |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 54 | | mean | 24.44 | | std | 18.19 | | cv | 0.744 | | sampleLengths | | 0 | 43 | | 1 | 14 | | 2 | 55 | | 3 | 8 | | 4 | 6 | | 5 | 44 | | 6 | 29 | | 7 | 23 | | 8 | 5 | | 9 | 46 | | 10 | 51 | | 11 | 15 | | 12 | 39 | | 13 | 19 | | 14 | 4 | | 15 | 15 | | 16 | 63 | | 17 | 19 | | 18 | 43 | | 19 | 43 | | 20 | 11 | | 21 | 67 | | 22 | 27 | | 23 | 13 | | 24 | 12 | | 25 | 31 | | 26 | 40 | | 27 | 5 | | 28 | 44 | | 29 | 16 | | 30 | 15 | | 31 | 33 | | 32 | 1 | | 33 | 41 | | 34 | 50 | | 35 | 5 | | 36 | 40 | | 37 | 47 | | 38 | 21 | | 39 | 28 | | 40 | 1 | | 41 | 5 | | 42 | 36 | | 43 | 6 | | 44 | 31 | | 45 | 50 | | 46 | 18 | | 47 | 4 | | 48 | 8 | | 49 | 8 |
| |
| 100.00% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 0 | | totalSentences | 99 | | matches | (empty) | |
| 40.32% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 4 | | totalVerbs | 167 | | matches | | 0 | "was manipulating" | | 1 | "was not lying" | | 2 | "was crossing" | | 3 | "was hunting" |
| |
| 97.14% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 2 | | semicolonCount | 1 | | flaggedSentences | 2 | | totalSentences | 125 | | ratio | 0.016 | | matches | | 0 | "He leaned lightly on an ivory-handled cane, his mismatched eyes—one the colour of aged amber, the other a pit of black—fixed on her." | | 1 | "Lucien’s world was not one of bruises and angry words; it was one of ancient blood and binding pacts." |
| |
| 94.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1037 | | adjectiveStacks | 1 | | stackExamples | | 0 | "faint, sickly purple light," |
| | adverbCount | 24 | | adverbRatio | 0.023143683702989394 | | lyAdverbCount | 10 | | lyAdverbRatio | 0.009643201542912247 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 125 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 125 | | mean | 10.56 | | std | 7.79 | | cv | 0.738 | | sampleLengths | | 0 | 23 | | 1 | 14 | | 2 | 6 | | 3 | 14 | | 4 | 21 | | 5 | 23 | | 6 | 8 | | 7 | 3 | | 8 | 8 | | 9 | 6 | | 10 | 23 | | 11 | 17 | | 12 | 4 | | 13 | 7 | | 14 | 16 | | 15 | 6 | | 16 | 23 | | 17 | 5 | | 18 | 12 | | 19 | 18 | | 20 | 16 | | 21 | 20 | | 22 | 30 | | 23 | 1 | | 24 | 11 | | 25 | 4 | | 26 | 10 | | 27 | 18 | | 28 | 11 | | 29 | 12 | | 30 | 4 | | 31 | 3 | | 32 | 4 | | 33 | 4 | | 34 | 11 | | 35 | 20 | | 36 | 2 | | 37 | 2 | | 38 | 21 | | 39 | 7 | | 40 | 11 | | 41 | 14 | | 42 | 5 | | 43 | 24 | | 44 | 12 | | 45 | 7 | | 46 | 10 | | 47 | 14 | | 48 | 19 | | 49 | 11 |
| |
| 48.27% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 13 | | diversityRatio | 0.352 | | totalSentences | 125 | | uniqueOpeners | 44 | |
| 72.46% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 2 | | totalSentences | 92 | | matches | | 0 | "Suddenly, the scent of ozone" | | 1 | "Then came the sound." |
| | ratio | 0.022 | |
| 20.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 46 | | totalSentences | 92 | | matches | | 0 | "He leaned lightly on an" | | 1 | "He did not look surprised" | | 2 | "He never did." | | 3 | "Her hand tightened on the" | | 4 | "He used the name she" | | 5 | "She tried to slam the" | | 6 | "His cane shot out, the" | | 7 | "His voice was calm, a" | | 8 | "He applied subtle, inexorable pressure," | | 9 | "He stepped inside, bringing with" | | 10 | "He moved with a liquid" | | 11 | "He cast a dismissive glance" | | 12 | "His gaze lingered on the" | | 13 | "It was not a question" | | 14 | "It was a verdict." | | 15 | "she corrected, her voice tight" | | 16 | "She closed the door, the" | | 17 | "He left no mark." | | 18 | "His amber eye flared." | | 19 | "She crossed her arms, a" |
| | ratio | 0.5 | |
| 0.00% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 85 | | totalSentences | 92 | | matches | | 0 | "The sharp knock rattled the" | | 1 | "Aurora yanked the door open," | | 2 | "The words died in her" | | 3 | "Lucien Moreau stood on the" | | 4 | "Rain slicked his platinum blond" | | 5 | "He leaned lightly on an" | | 6 | "He did not look surprised" | | 7 | "He never did." | | 8 | "Her hand tightened on the" | | 9 | "He used the name she" | | 10 | "A faint smile touched his" | | 11 | "She tried to slam the" | | 12 | "His cane shot out, the" | | 13 | "The impact shuddered up her" | | 14 | "His voice was calm, a" | | 15 | "He applied subtle, inexorable pressure," | | 16 | "He stepped inside, bringing with" | | 17 | "He moved with a liquid" | | 18 | "He cast a dismissive glance" | | 19 | "His gaze lingered on the" |
| | ratio | 0.924 | |
| 54.35% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 1 | | totalSentences | 92 | | matches | | 0 | "If you wanted something found," |
| | ratio | 0.011 | |
| 32.97% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 39 | | technicalSentenceCount | 6 | | matches | | 0 | "He moved with a liquid grace that was utterly at odds with the cramped, cluttered flat." | | 1 | "His gaze lingered on the worn kilim rug, the half-eaten bowl of noodles on the coffee table, the tabby cat, Ptolemy, who was now winding around his legs, purrin…" | | 2 | "His eyes lingered on it as if it were a brand." | | 3 | "His touch was cold, a shock against her skin that jolted down her spine." | | 4 | "The beautiful, terrifying man who saw the world in terms of ownership, and who had made it devastatingly clear that he considered her his most prized possession…" | | 5 | "It was a smell from her darkest childhood memories, a smell that meant something was crossing over." |
| |
| 17.86% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 14 | | uselessAdditionCount | 3 | | matches | | 0 | "she corrected, her voice tight" | | 1 | "he continued, his voice dropping to a low murmur" | | 2 | "he said, his voice dangerously soft," |
| |
| 80.23% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 4 | | fancyCount | 3 | | fancyTags | | 0 | "she corrected (correct)" | | 1 | "she whispered (whisper)" | | 2 | "he continued (continue)" |
| | dialogueSentences | 43 | | tagDensity | 0.093 | | leniency | 0.186 | | rawRatio | 0.75 | | effectiveRatio | 0.14 | |