| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 3 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 7 | | tagDensity | 0.429 | | leniency | 0.857 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 90.57% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1591 | | totalAiIsmAdverbs | 3 | | found | | | highlights | | 0 | "slowly" | | 1 | "really" | | 2 | "loosely" |
| |
| 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) | |
| 49.72% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1591 | | totalAiIsms | 16 | | found | | 0 | | | 1 | | | 2 | | | 3 | | | 4 | | word | "down her spine" | | count | 1 |
| | 5 | | | 6 | | | 7 | | | 8 | | | 9 | | | 10 | | | 11 | | | 12 | | | 13 | | | 14 | |
| | highlights | | 0 | "trembled" | | 1 | "furrowed" | | 2 | "dance" | | 3 | "chill" | | 4 | "down her spine" | | 5 | "racing" | | 6 | "reverberated" | | 7 | "pulsed" | | 8 | "absolutely" | | 9 | "beacon" | | 10 | "etching" | | 11 | "weight" | | 12 | "glinting" | | 13 | "clenching" | | 14 | "ceaseless" |
| |
| 100.00% | Cliché density | Target: ≤1 cliche(s) per 800-word window | | totalCliches | 1 | | maxInWindow | 1 | | found | | 0 | | label | "sent a shiver through" | | count | 1 |
|
| | highlights | | |
| 100.00% | Emotion telling (show vs. tell) | Target: ≤3% sentences with emotion telling | | emotionTells | 1 | | narrationSentences | 115 | | matches | | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 1 | | hedgeCount | 2 | | narrationSentences | 115 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 119 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 72 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 0 | | markdownWords | 0 | | totalWords | 1572 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 4 | | unquotedAttributions | 0 | | matches | (empty) | |
| 50.00% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 44 | | wordCount | 1550 | | uniqueNames | 20 | | maxNameDensity | 1.16 | | worstName | "Quinn" | | maxWindowNameDensity | 3.5 | | worstWindowName | "Quinn" | | discoveredNames | | Harlow | 1 | | Quinn | 18 | | London | 1 | | English | 1 | | Setup | 1 | | Metropolitan | 1 | | Police | 1 | | Detective | 2 | | Tube | 1 | | Reserve | 1 | | Tomas | 7 | | Master | 1 | | Samson | 1 | | Achilles | 1 | | Chinese | 1 | | Khwis | 1 | | Origins | 1 | | Armii | 1 | | Surrounded | 1 | | European | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Police" | | 3 | "Detective" | | 4 | "Tomas" | | 5 | "Master" | | 6 | "Achilles" | | 7 | "Chinese" |
| | places | | | globalScore | 0.919 | | windowScore | 0.5 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 85 | | 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 | 1572 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 119 | | matches | (empty) | |
| 96.51% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 34 | | mean | 46.24 | | std | 22.55 | | cv | 0.488 | | sampleLengths | | 0 | 47 | | 1 | 54 | | 2 | 28 | | 3 | 35 | | 4 | 3 | | 5 | 23 | | 6 | 22 | | 7 | 8 | | 8 | 11 | | 9 | 25 | | 10 | 31 | | 11 | 37 | | 12 | 43 | | 13 | 24 | | 14 | 56 | | 15 | 30 | | 16 | 43 | | 17 | 46 | | 18 | 57 | | 19 | 54 | | 20 | 87 | | 21 | 50 | | 22 | 44 | | 23 | 87 | | 24 | 65 | | 25 | 76 | | 26 | 78 | | 27 | 36 | | 28 | 97 | | 29 | 60 | | 30 | 66 | | 31 | 37 | | 32 | 48 | | 33 | 64 |
| |
| 86.96% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 6 | | totalSentences | 115 | | matches | | 0 | "was dammed" | | 1 | "were worn" | | 2 | "was busted" | | 3 | "were banned" | | 4 | "was joined" | | 5 | "are conveyed" |
| |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 2 | | totalVerbs | 250 | | matches | | 0 | "was shaking" | | 1 | "were reeling" |
| |
| 94.84% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 3 | | semicolonCount | 0 | | flaggedSentences | 2 | | totalSentences | 119 | | ratio | 0.017 | | matches | | 0 | "Stray words slip out onto the lingering shadows and delicious odors from loosely barricade peek holes, the shadowy drifts under the sticky and ageless virtual object illustrations - the grinding ale and-centimeter small pebbles gutted according to the stone coast." | | 1 | "She follows the glimmering device into an alcove where Tomas connects with a long syllables that have been cut, cutting the high-wire act into cables - a set of looking like just catcher fingers of a cuff that would be rekindling on the inside - the honey crushed all at a sudden." |
| |
| 88.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1568 | | adjectiveStacks | 2 | | stackExamples | | 0 | "next dew-cloaked apples," | | 1 | "unkindly intact chemical hydrant" |
| | adverbCount | 38 | | adverbRatio | 0.02423469387755102 | | lyAdverbCount | 14 | | lyAdverbRatio | 0.008928571428571428 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 119 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 119 | | mean | 13.21 | | std | 12.52 | | cv | 0.948 | | sampleLengths | | 0 | 13 | | 1 | 13 | | 2 | 21 | | 3 | 6 | | 4 | 3 | | 5 | 12 | | 6 | 6 | | 7 | 12 | | 8 | 15 | | 9 | 15 | | 10 | 9 | | 11 | 4 | | 12 | 13 | | 13 | 6 | | 14 | 16 | | 15 | 3 | | 16 | 13 | | 17 | 8 | | 18 | 2 | | 19 | 3 | | 20 | 4 | | 21 | 3 | | 22 | 3 | | 23 | 2 | | 24 | 3 | | 25 | 4 | | 26 | 5 | | 27 | 3 | | 28 | 4 | | 29 | 7 | | 30 | 9 | | 31 | 6 | | 32 | 4 | | 33 | 2 | | 34 | 4 | | 35 | 4 | | 36 | 16 | | 37 | 6 | | 38 | 5 | | 39 | 6 | | 40 | 10 | | 41 | 7 | | 42 | 10 | | 43 | 2 | | 44 | 2 | | 45 | 8 | | 46 | 5 | | 47 | 3 | | 48 | 4 | | 49 | 4 |
| |
| 85.99% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 6 | | diversityRatio | 0.5462184873949579 | | totalSentences | 119 | | uniqueOpeners | 65 | |
| 100.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 5 | | totalSentences | 110 | | matches | | 0 | "Definitely not English." | | 1 | "Maybe she had him." | | 2 | "Slowly, she glanced over her" | | 3 | "Just had to keep low" | | 4 | "Then a snatch of stores" |
| | ratio | 0.045 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 19 | | totalSentences | 110 | | matches | | 0 | "She couldn't let him get" | | 1 | "She spotted her quarry ducking" | | 2 | "He held up his hands" | | 3 | "She didn't recognize him." | | 4 | "He was dark-complected." | | 5 | "It was a Setup." | | 6 | "She could just make out" | | 7 | "Her partner's blood on her" | | 8 | "Her sidearm would be about" | | 9 | "She'd be on her own" | | 10 | "Her senses were reeling." | | 11 | "Her eyes kept flitting to" | | 12 | "Her finger fingered the commentary" | | 13 | "She was so far from" | | 14 | "She fixed her gaze on" | | 15 | "Their voices she hears but" | | 16 | "She follows Tomas between stalls" | | 17 | "He winding through the tower" | | 18 | "She follows the glimmering device" |
| | ratio | 0.173 | |
| 100.00% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 77 | | totalSentences | 110 | | matches | | 0 | "Detective Harlow Quinn's shoes splashed" | | 1 | "The narrow streets of London" | | 2 | "The warm blood rushed in" | | 3 | "She couldn't let him get" | | 4 | "The missing piece of the" | | 5 | "The answer to the question" | | 6 | "Quinn's lungs burned as she" | | 7 | "She spotted her quarry ducking" | | 8 | "The man skidded to a" | | 9 | "Quinn took advantage, closing the" | | 10 | "The man spun on his" | | 11 | "He held up his hands" | | 12 | "Quinn's brow furrowed." | | 13 | "She didn't recognize him." | | 14 | "He was dark-complected." | | 15 | "Olympus, more like it." | | 16 | "the man stammered" | | 17 | "Quinn glowered at him, finger" | | 18 | "Tomas was shaking like a" | | 19 | "It was a Setup." |
| | ratio | 0.7 | |
| 90.91% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 2 | | totalSentences | 110 | | matches | | 0 | "If they found out she" | | 1 | "If she lost, so too" |
| | ratio | 0.018 | |
| 96.77% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 62 | | technicalSentenceCount | 4 | | matches | | 0 | "Quinn kept Tomas in her sight as she followed him past the tables laden with fruit jars all aglow with an inner light, textbooks with toll marks, pendulums that…" | | 1 | "Her eyes kept flitting to the impossible flasks on display in blown grey glass that allowed buyers to sample their wares thoughtlessly." | | 2 | "She follows the glimmering device into an alcove where Tomas connects with a long syllables that have been cut, cutting the high-wire act into cables - a set of…" | | 3 | "Blistered and weeping, ample puddles of barbecued backwaters of strategic patients, the one to the mirth of thanatized swarms glaring overhangs, the next dew-cl…" |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 3 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 0.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 2 | | fancyCount | 2 | | fancyTags | | 0 | "Quinn demanded (demand)" | | 1 | "the man stammered (stammer)" |
| | dialogueSentences | 7 | | tagDensity | 0.286 | | leniency | 0.571 | | rawRatio | 1 | | effectiveRatio | 0.571 | |