| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 7 | | adverbTagCount | 1 | | adverbTags | | 0 | "Eva's voice trembled slightly [slightly]" |
| | dialogueSentences | 36 | | tagDensity | 0.194 | | leniency | 0.389 | | rawRatio | 0.143 | | effectiveRatio | 0.056 | |
| 84.14% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 946 | | totalAiIsmAdverbs | 3 | | 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) | |
| 57.72% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 946 | | totalAiIsms | 8 | | found | | | highlights | | 0 | "pulsed" | | 1 | "warmth" | | 2 | "echoing" | | 3 | "etched" | | 4 | "familiar" | | 5 | "trembled" | | 6 | "disrupt" |
| |
| 66.67% | Cliché density | Target: ≤1 cliche(s) per 800-word window | | totalCliches | 2 | | maxInWindow | 2 | | found | | 0 | | label | "eyes widened/narrowed" | | count | 2 |
|
| | highlights | | |
| 100.00% | Emotion telling (show vs. tell) | Target: ≤3% sentences with emotion telling | | emotionTells | 0 | | narrationSentences | 57 | | matches | (empty) | |
| 92.73% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 1 | | hedgeCount | 1 | | narrationSentences | 57 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 86 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 30 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 0 | | markdownWords | 0 | | totalWords | 940 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 7 | | unquotedAttributions | 0 | | matches | (empty) | |
| 0.00% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 47 | | wordCount | 580 | | uniqueNames | 11 | | maxNameDensity | 2.76 | | worstName | "Eva" | | maxWindowNameDensity | 5.5 | | worstWindowName | "Eva" | | discoveredNames | | Quinn | 1 | | Tube | 2 | | Veil | 1 | | Market | 1 | | Morris | 6 | | Eva | 16 | | Kowalski | 1 | | British | 1 | | Museum | 1 | | Harlow | 16 | | London | 1 |
| | persons | | 0 | "Quinn" | | 1 | "Morris" | | 2 | "Eva" | | 3 | "Kowalski" | | 4 | "Harlow" |
| | places | | 0 | "Tube" | | 1 | "British" | | 2 | "Museum" | | 3 | "London" |
| | globalScore | 0.121 | | windowScore | 0 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 43 | | 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 | 940 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 86 | | matches | (empty) | |
| 76.52% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 38 | | mean | 24.74 | | std | 10.34 | | cv | 0.418 | | sampleLengths | | 0 | 48 | | 1 | 40 | | 2 | 40 | | 3 | 28 | | 4 | 32 | | 5 | 35 | | 6 | 19 | | 7 | 10 | | 8 | 20 | | 9 | 18 | | 10 | 22 | | 11 | 8 | | 12 | 36 | | 13 | 15 | | 14 | 16 | | 15 | 10 | | 16 | 31 | | 17 | 14 | | 18 | 30 | | 19 | 17 | | 20 | 18 | | 21 | 33 | | 22 | 21 | | 23 | 14 | | 24 | 11 | | 25 | 34 | | 26 | 29 | | 27 | 12 | | 28 | 18 | | 29 | 19 | | 30 | 28 | | 31 | 32 | | 32 | 26 | | 33 | 14 | | 34 | 38 | | 35 | 40 | | 36 | 37 | | 37 | 27 |
| |
| 86.80% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 3 | | totalSentences | 57 | | matches | | 0 | "was supposed" | | 1 | "were clenched" | | 2 | "was connected" |
| |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 1 | | totalVerbs | 95 | | matches | | |
| 43.19% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 3 | | semicolonCount | 0 | | flaggedSentences | 3 | | totalSentences | 86 | | ratio | 0.035 | | matches | | 0 | "\"Detective Quinn.\" The voice came from the shadows—Eva Kowalski, the research assistant from the British Museum." | | 1 | "They were clenched around something—a small brass compass, its face etched with protective sigils." | | 2 | "Harlow noticed something else—a trail of fine white powder leading from the altar to a crack in the wall." |
| |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 585 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 14 | | adverbRatio | 0.023931623931623933 | | lyAdverbCount | 5 | | lyAdverbRatio | 0.008547008547008548 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 86 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 86 | | mean | 10.93 | | std | 6.13 | | cv | 0.561 | | sampleLengths | | 0 | 23 | | 1 | 25 | | 2 | 10 | | 3 | 16 | | 4 | 14 | | 5 | 16 | | 6 | 11 | | 7 | 13 | | 8 | 16 | | 9 | 7 | | 10 | 5 | | 11 | 11 | | 12 | 21 | | 13 | 13 | | 14 | 14 | | 15 | 8 | | 16 | 8 | | 17 | 11 | | 18 | 4 | | 19 | 6 | | 20 | 13 | | 21 | 7 | | 22 | 9 | | 23 | 9 | | 24 | 10 | | 25 | 12 | | 26 | 6 | | 27 | 2 | | 28 | 6 | | 29 | 17 | | 30 | 13 | | 31 | 8 | | 32 | 7 | | 33 | 4 | | 34 | 8 | | 35 | 4 | | 36 | 5 | | 37 | 5 | | 38 | 19 | | 39 | 8 | | 40 | 4 | | 41 | 3 | | 42 | 11 | | 43 | 9 | | 44 | 12 | | 45 | 9 | | 46 | 3 | | 47 | 14 | | 48 | 10 | | 49 | 8 |
| |
| 58.14% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 0 | | diversityRatio | 0.3488372093023256 | | totalSentences | 86 | | uniqueOpeners | 30 | |
| 60.61% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 1 | | totalSentences | 55 | | matches | | 0 | "Somewhere in London, another rift" |
| | ratio | 0.018 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 11 | | totalSentences | 55 | | matches | | 0 | "She descended the rusted stairs," | | 1 | "Her partner, DS Morris, lay" | | 2 | "She emerged from the darkness," | | 3 | "She kept her voice steady" | | 4 | "She gestured to a rusted" | | 5 | "They were clenched around something—a" | | 6 | "She knelt, touching it with" | | 7 | "She pulled out a heavy" | | 8 | "She emerged from the Tube" | | 9 | "She'd been solving impossible cases" | | 10 | "She pulled out her phone" |
| | ratio | 0.2 | |
| 14.55% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 49 | | totalSentences | 55 | | matches | | 0 | "Harlow Quinn stood at the" | | 1 | "The Veil Market was supposed" | | 2 | "She descended the rusted stairs," | | 3 | "The platform stretched before her," | | 4 | "Her partner, DS Morris, lay" | | 5 | "The voice came from the" | | 6 | "She emerged from the darkness," | | 7 | "Harlow approached Morris's body, noting" | | 8 | "She kept her voice steady" | | 9 | "Eva tucked a curl of" | | 10 | "She gestured to a rusted" | | 11 | "Harlow knelt beside Morris, her" | | 12 | "They were clenched around something—a" | | 13 | "The needle spun wildly, pointing" | | 14 | "Eva's voice held a note" | | 15 | "Harlow pocketed the compass." | | 16 | "Eva's eyes darted to the" | | 17 | "Harlow stood, her military precision" | | 18 | "Eva nodded, her freckled complexion" | | 19 | "Harlow turned to the metal" |
| | ratio | 0.891 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 55 | | matches | (empty) | | ratio | 0 | |
| 100.00% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 27 | | technicalSentenceCount | 1 | | matches | | 0 | "The Veil Market was supposed to be empty tonight, but the bone token in her pocket pulsed with an unnatural warmth that told her otherwise." |
| |
| 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 | 1 | | fancyCount | 1 | | fancyTags | | 0 | "Eva explained (explain)" |
| | dialogueSentences | 36 | | tagDensity | 0.028 | | leniency | 0.056 | | rawRatio | 1 | | effectiveRatio | 0.056 | |