| 25.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 20 | | adverbTagCount | 5 | | adverbTags | | 0 | "She gestured vaguely [vaguely]" | | 1 | "His voice softened slightly [slightly]" | | 2 | "She looked away [away]" | | 3 | "He gestured vaguely [vaguely]" | | 4 | "she said softly [softly]" |
| | dialogueSentences | 31 | | tagDensity | 0.645 | | leniency | 1 | | rawRatio | 0.25 | | effectiveRatio | 0.25 | |
| 71.99% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1071 | | totalAiIsmAdverbs | 6 | | found | | 0 | | | 1 | | | 2 | | | 3 | | adverb | "barely above a whisper" | | count | 1 |
|
| | highlights | | 0 | "softly" | | 1 | "slightly" | | 2 | "carefully" | | 3 | "barely above a whisper" |
| |
| 100.00% | AI-ism character names | Target: 0 AI-default names (16 tracked, −20% each) | | codexExemptions | | | found | (empty) | |
| 100.00% | AI-ism location names | Target: 0 AI-default location names (33 tracked, −20% each) | | codexExemptions | (empty) | | found | (empty) | |
| 0.00% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1071 | | totalAiIsms | 25 | | found | | | highlights | | 0 | "perfect" | | 1 | "sanctuary" | | 2 | "familiar" | | 3 | "silence" | | 4 | "weight" | | 5 | "warmth" | | 6 | "scanned" | | 7 | "unspoken" | | 8 | "unreadable" | | 9 | "pulse" | | 10 | "shattered" | | 11 | "whisper" | | 12 | "stark" | | 13 | "reminder" |
| |
| 100.00% | Cliché density | Target: ≤1 cliche(s) per 800-word window | | totalCliches | 1 | | maxInWindow | 1 | | found | | 0 | | label | "hung in the air" | | count | 1 |
|
| | highlights | | |
| 100.00% | Emotion telling (show vs. tell) | Target: ≤3% sentences with emotion telling | | emotionTells | 0 | | narrationSentences | 65 | | matches | (empty) | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 1 | | narrationSentences | 65 | | filterMatches | (empty) | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 71 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 32 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 2 | | markdownWords | 9 | | totalWords | 1069 | | ratio | 0.008 | | matches | | 0 | "What did you become?" | | 1 | "Why didn’t you tell me?" |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 10 | | unquotedAttributions | 0 | | matches | (empty) | |
| 100.00% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 21 | | wordCount | 815 | | uniqueNames | 10 | | maxNameDensity | 0.74 | | worstName | "Silas" | | maxWindowNameDensity | 1.5 | | worstWindowName | "Silas" | | discoveredNames | | Raven | 1 | | Nest | 1 | | Carter | 1 | | Rory | 1 | | Laila | 1 | | London | 2 | | Silas | 6 | | Blackwood | 2 | | Prague | 1 | | Aurora | 5 |
| | persons | | 0 | "Raven" | | 1 | "Nest" | | 2 | "Carter" | | 3 | "Rory" | | 4 | "Laila" | | 5 | "Silas" | | 6 | "Blackwood" | | 7 | "Aurora" |
| | places | | | globalScore | 1 | | windowScore | 1 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 56 | | 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 | 1069 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 71 | | matches | (empty) | |
| 48.07% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 14 | | mean | 76.36 | | std | 24.3 | | cv | 0.318 | | sampleLengths | | 0 | 116 | | 1 | 94 | | 2 | 82 | | 3 | 111 | | 4 | 31 | | 5 | 57 | | 6 | 45 | | 7 | 64 | | 8 | 63 | | 9 | 65 | | 10 | 95 | | 11 | 89 | | 12 | 98 | | 13 | 59 |
| |
| 100.00% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 0 | | totalSentences | 65 | | matches | (empty) | |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 0 | | totalVerbs | 122 | | matches | (empty) | |
| 0.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 2 | | semicolonCount | 2 | | flaggedSentences | 4 | | totalSentences | 71 | | ratio | 0.056 | | matches | | 0 | "The Raven’s Nest smelled of aged oak, spilled beer, and something sharper—like ozone and old secrets." | | 1 | "Silas, who’d been her mentor, her anchor, her friend—before the Prague operation, before the limp, before the silence." | | 2 | "He didn’t sit; he leaned against the bar, his back to her, his eyes fixed on the swirling amber liquid in his glass." | | 3 | "The silence wasn’t awkward anymore; it was a shared understanding, a bridge built on shattered pieces." |
| |
| 91.79% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 817 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 34 | | adverbRatio | 0.0416156670746634 | | lyAdverbCount | 23 | | lyAdverbRatio | 0.028151774785801713 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 71 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 71 | | mean | 15.06 | | std | 7.24 | | cv | 0.481 | | sampleLengths | | 0 | 16 | | 1 | 23 | | 2 | 10 | | 3 | 24 | | 4 | 17 | | 5 | 25 | | 6 | 1 | | 7 | 10 | | 8 | 11 | | 9 | 11 | | 10 | 6 | | 11 | 22 | | 12 | 3 | | 13 | 16 | | 14 | 15 | | 15 | 5 | | 16 | 15 | | 17 | 4 | | 18 | 12 | | 19 | 18 | | 20 | 18 | | 21 | 10 | | 22 | 16 | | 23 | 25 | | 24 | 12 | | 25 | 14 | | 26 | 23 | | 27 | 21 | | 28 | 8 | | 29 | 9 | | 30 | 10 | | 31 | 4 | | 32 | 3 | | 33 | 10 | | 34 | 20 | | 35 | 18 | | 36 | 6 | | 37 | 21 | | 38 | 18 | | 39 | 6 | | 40 | 12 | | 41 | 24 | | 42 | 19 | | 43 | 9 | | 44 | 10 | | 45 | 32 | | 46 | 21 | | 47 | 10 | | 48 | 20 | | 49 | 12 |
| |
| 66.20% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 3 | | diversityRatio | 0.4225352112676056 | | totalSentences | 71 | | uniqueOpeners | 30 | |
| 52.08% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 1 | | totalSentences | 64 | | matches | | 0 | "Then, the door chimed again." |
| | ratio | 0.016 | |
| 32.50% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 30 | | totalSentences | 64 | | matches | | 0 | "She’d delivered to this bar" | | 1 | "She spotted a vacant stool" | | 2 | "She slid onto the stool," | | 3 | "She watched him for a" | | 4 | "She took a slow sip," | | 5 | "He didn’t move immediately, just" | | 6 | "His right hand rested near" | | 7 | "He didn’t sit; he leaned" | | 8 | "She took another sip, forcing" | | 9 | "Her voice was low, devoid" | | 10 | "He didn’t turn." | | 11 | "His voice was gravelly, a" | | 12 | "He finally shifted, turning his" | | 13 | "It wasn’t a question" | | 14 | "she countered, her gaze flicking" | | 15 | "He took a slow sip" | | 16 | "He finally slid onto the" | | 17 | "His eyes scanned her face," | | 18 | "She gestured vaguely towards the" | | 19 | "He finally met her eyes" |
| | ratio | 0.469 | |
| 0.00% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 61 | | totalSentences | 64 | | matches | | 0 | "The Raven’s Nest smelled of" | | 1 | "Aurora Carter, alias Rory, alias" | | 2 | "She’d delivered to this bar" | | 3 | "Tonight, she was here for" | | 4 | "The bar was a labyrinth" | | 5 | "She spotted a vacant stool" | | 6 | "She slid onto the stool," | | 7 | "The bartender, a man with" | | 8 | "Aurora handed over a ten-pound" | | 9 | "The bartender’s movements were precise," | | 10 | "She watched him for a" | | 11 | "A veteran, probably." | | 12 | "She took a slow sip," | | 13 | "The bar was her sanctuary" | | 14 | "A familiar scent, like damp" | | 15 | "Aurora’s head snapped up." | | 16 | "Silas, who’d been her mentor," | | 17 | "The man who’d taught her" | | 18 | "The man who’d vanished without" | | 19 | "Silas’s gaze swept the bar," |
| | ratio | 0.953 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 64 | | matches | (empty) | | ratio | 0 | |
| 71.43% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 30 | | technicalSentenceCount | 3 | | matches | | 0 | "Silas, who’d been her mentor, her anchor, her friend—before the Prague operation, before the limp, before the silence." | | 1 | "The man who’d taught her to read people like open books, to trust her instincts above all else." | | 2 | "Silas’s gaze swept the bar, landing on her with a precision that made her skin prickle." |
| |
| 75.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 20 | | uselessAdditionCount | 2 | | matches | | 0 | "she admitted, the words tumbling out before she could stop them" | | 1 | "she asked, her voice barely above a whisper" |
| |
| 85.48% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 4 | | fancyCount | 2 | | fancyTags | | 0 | "the leather groaning (groan)" | | 1 | "she admitted (admit)" |
| | dialogueSentences | 31 | | tagDensity | 0.129 | | leniency | 0.258 | | rawRatio | 0.5 | | effectiveRatio | 0.129 | |