| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 4 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 24 | | tagDensity | 0.167 | | leniency | 0.333 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 91.73% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1209 | | totalAiIsmAdverbs | 2 | | 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) | |
| 71.05% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1209 | | totalAiIsms | 7 | | found | | | highlights | | 0 | "familiar" | | 1 | "weight" | | 2 | "gloom" | | 3 | "echo" | | 4 | "predator" | | 5 | "scanned" | | 6 | "velvet" |
| |
| 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 | 0 | | narrationSentences | 125 | | matches | (empty) | |
| 85.71% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 3 | | hedgeCount | 2 | | narrationSentences | 125 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 145 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 42 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 4 | | markdownWords | 9 | | totalWords | 1204 | | ratio | 0.007 | | matches | | 0 | "Herrera" | | 1 | "What are you doing here, Harlow?" | | 2 | "Movement." | | 3 | "supernatural" |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 5 | | unquotedAttributions | 0 | | matches | (empty) | |
| 100.00% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 8 | | wordCount | 1012 | | uniqueNames | 6 | | maxNameDensity | 0.2 | | worstName | "Tommy" | | maxWindowNameDensity | 1 | | worstWindowName | "Yard" | | discoveredNames | | Soho | 1 | | Veil | 1 | | Market | 1 | | Tommy | 2 | | Controlled | 1 | | Yard | 2 |
| | persons | | | places | | | globalScore | 1 | | windowScore | 1 | |
| 45.83% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 72 | | glossingSentenceCount | 3 | | matches | | 0 | "looked like inscriptions, scratched deep" | | 1 | "looked like they’d been cobbled together" | | 2 | "looked like a rotting corpse dressed in v" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 1 | | per1kWords | 0.831 | | wordCount | 1204 | | matches | | 0 | "Not animals, but things" |
| |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 1 | | totalSentences | 145 | | matches | | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 61 | | mean | 19.74 | | std | 16.26 | | cv | 0.824 | | sampleLengths | | 0 | 60 | | 1 | 48 | | 2 | 46 | | 3 | 28 | | 4 | 1 | | 5 | 35 | | 6 | 36 | | 7 | 8 | | 8 | 2 | | 9 | 66 | | 10 | 29 | | 11 | 15 | | 12 | 33 | | 13 | 3 | | 14 | 29 | | 15 | 53 | | 16 | 6 | | 17 | 23 | | 18 | 6 | | 19 | 1 | | 20 | 55 | | 21 | 31 | | 22 | 3 | | 23 | 8 | | 24 | 32 | | 25 | 1 | | 26 | 20 | | 27 | 16 | | 28 | 15 | | 29 | 6 | | 30 | 2 | | 31 | 7 | | 32 | 14 | | 33 | 28 | | 34 | 1 | | 35 | 5 | | 36 | 2 | | 37 | 6 | | 38 | 14 | | 39 | 16 | | 40 | 4 | | 41 | 6 | | 42 | 17 | | 43 | 36 | | 44 | 3 | | 45 | 7 | | 46 | 16 | | 47 | 42 | | 48 | 16 | | 49 | 14 |
| |
| 94.04% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 4 | | totalSentences | 125 | | matches | | 0 | "was stuck" | | 1 | "was supposed" | | 2 | "was revoked" | | 3 | "were lined" | | 4 | "been cobbled" |
| |
| 49.34% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 4 | | totalVerbs | 177 | | matches | | 0 | "was hawking" | | 1 | "was buying" | | 2 | "was watching" | | 3 | "was sharpening" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 1 | | semicolonCount | 1 | | flaggedSentences | 2 | | totalSentences | 145 | | ratio | 0.014 | | matches | | 0 | "It had stopped working years ago; the second hand was stuck at four minutes past midnight, a permanent marker of the night my partner died." | | 1 | "The smell hit me first—musty earth mixed with the sharp, metallic tang of ozone, like high-voltage electricity burning through dust." |
| |
| 88.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1017 | | adjectiveStacks | 2 | | stackExamples | | 0 | "Under harsh, unnatural light," | | 1 | "small, bone-white token" |
| | adverbCount | 34 | | adverbRatio | 0.03343166175024582 | | lyAdverbCount | 8 | | lyAdverbRatio | 0.007866273352999017 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 145 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 145 | | mean | 8.3 | | std | 6.24 | | cv | 0.752 | | sampleLengths | | 0 | 11 | | 1 | 3 | | 2 | 11 | | 3 | 18 | | 4 | 17 | | 5 | 3 | | 6 | 12 | | 7 | 25 | | 8 | 5 | | 9 | 1 | | 10 | 1 | | 11 | 1 | | 12 | 8 | | 13 | 12 | | 14 | 5 | | 15 | 4 | | 16 | 17 | | 17 | 14 | | 18 | 4 | | 19 | 10 | | 20 | 1 | | 21 | 8 | | 22 | 1 | | 23 | 9 | | 24 | 9 | | 25 | 8 | | 26 | 10 | | 27 | 3 | | 28 | 12 | | 29 | 11 | | 30 | 7 | | 31 | 1 | | 32 | 2 | | 33 | 10 | | 34 | 11 | | 35 | 7 | | 36 | 6 | | 37 | 20 | | 38 | 3 | | 39 | 9 | | 40 | 5 | | 41 | 15 | | 42 | 9 | | 43 | 15 | | 44 | 3 | | 45 | 3 | | 46 | 19 | | 47 | 8 | | 48 | 3 | | 49 | 5 |
| |
| 48.51% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 18 | | diversityRatio | 0.36551724137931035 | | totalSentences | 145 | | uniqueOpeners | 53 | |
| 89.29% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 3 | | totalSentences | 112 | | matches | | 0 | "Then the whispers." | | 1 | "Especially men whose partners died" | | 2 | "Then I looked at the" |
| | ratio | 0.027 | |
| 16.43% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 57 | | totalSentences | 112 | | matches | | 0 | "I didn’t cough." | | 1 | "My lungs held, burning with" | | 2 | "He scrambled over the brick" | | 3 | "I didn't hesitate." | | 4 | "I vaulted the fence myself," | | 5 | "It had stopped working years" | | 6 | "I dropped to the other" | | 7 | "He knew the score." | | 8 | "He cut down an alleyway" | | 9 | "He was fast, but then" | | 10 | "He was supposed to be" | | 11 | "His license was revoked, but" | | 12 | "I skidded around the corner," | | 13 | "I holstered my .38, checked" | | 14 | "My breath fogged in front" | | 15 | "I switched on my torch." | | 16 | "It looked like inscriptions, scratched" | | 17 | "I stepped deeper." | | 18 | "I kept my hands visible." | | 19 | "I walked past a stall" |
| | ratio | 0.509 | |
| 53.75% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 91 | | totalSentences | 112 | | matches | | 0 | "The exhaust fumes of a" | | 1 | "I didn’t cough." | | 2 | "My lungs held, burning with" | | 3 | "The fleeing figure ahead vaulted" | | 4 | "He scrambled over the brick" | | 5 | "I didn't hesitate." | | 6 | "I vaulted the fence myself," | | 7 | "It had stopped working years" | | 8 | "Rain hacked at my face," | | 9 | "I dropped to the other" | | 10 | "The man didn't look back." | | 11 | "He knew the score." | | 12 | "A quick sprint through Soho," | | 13 | "He cut down an alleyway" | | 14 | "He was fast, but then" | | 15 | "The name hit me like" | | 16 | "He was supposed to be" | | 17 | "His license was revoked, but" | | 18 | "I skidded around the corner," | | 19 | "The alley dead-ended." |
| | ratio | 0.813 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 112 | | matches | (empty) | | ratio | 0 | |
| 53.57% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 40 | | technicalSentenceCount | 5 | | matches | | 0 | "The beam cut a thin yellow path through the gloom, illuminating graffiti that wasn't paint." | | 1 | "The walls were lined with stalls, crude wooden structures that looked like they’d been cobbled together from shipping crates." | | 2 | "Beside him, a woman with skin the texture of parchment was hawking vials of liquid that glowed with an inner fire." | | 3 | "He pressed a small, bone-white token onto the counter of a vendor that looked like a rotting corpse dressed in velvet." | | 4 | "Especially men whose partners died under unexplained circumstances that were *supernatural*." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 4 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 3 | | fancyCount | 0 | | fancyTags | (empty) | | dialogueSentences | 24 | | tagDensity | 0.125 | | leniency | 0.25 | | rawRatio | 0 | | effectiveRatio | 0 | |