| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 14 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 53 | | tagDensity | 0.264 | | leniency | 0.528 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 80.51% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1283 | | totalAiIsmAdverbs | 5 | | found | | | highlights | | 0 | "gently" | | 1 | "lazily" | | 2 | "slowly" | | 3 | "carefully" | | 4 | "very" |
| |
| 60.00% | AI-ism character names | Target: 0 AI-default names (17 tracked, −20% each) | | codexExemptions | (empty) | | found | | |
| 100.00% | AI-ism location names | Target: 0 AI-default location names (33 tracked, −20% each) | | codexExemptions | (empty) | | found | (empty) | |
| 84.41% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1283 | | totalAiIsms | 4 | | found | | | highlights | | 0 | "velvet" | | 1 | "etched" | | 2 | "database" | | 3 | "eyebrow" |
| |
| 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 | 75 | | matches | (empty) | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 1 | | hedgeCount | 1 | | narrationSentences | 75 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 113 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 46 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 1 | | markdownWords | 13 | | totalWords | 1276 | | ratio | 0.01 | | matches | | 0 | "The market moves tomorrow night. Full moon. If you want answers, come alone." |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 13 | | unquotedAttributions | 0 | | matches | (empty) | |
| 56.13% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 40 | | wordCount | 799 | | uniqueNames | 14 | | maxNameDensity | 1.88 | | worstName | "Quinn" | | maxWindowNameDensity | 3 | | worstWindowName | "Quinn" | | discoveredNames | | Harlow | 1 | | Quinn | 15 | | Chen | 5 | | Veil | 1 | | Market | 1 | | Tube | 1 | | Camden | 2 | | Webb | 3 | | Eva | 6 | | Marcus | 1 | | Morris | 1 | | High | 1 | | Street | 1 | | London | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Chen" | | 3 | "Webb" | | 4 | "Eva" | | 5 | "Marcus" | | 6 | "Morris" |
| | places | | 0 | "Camden" | | 1 | "High" | | 2 | "Street" | | 3 | "London" |
| | globalScore | 0.561 | | windowScore | 0.667 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 54 | | glossingSentenceCount | 1 | | matches | | 0 | "things that seemed to shift when viewed from different angles" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 1276 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 113 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 64 | | mean | 19.94 | | std | 17.42 | | cv | 0.874 | | sampleLengths | | 0 | 47 | | 1 | 33 | | 2 | 58 | | 3 | 32 | | 4 | 14 | | 5 | 41 | | 6 | 34 | | 7 | 31 | | 8 | 3 | | 9 | 21 | | 10 | 17 | | 11 | 41 | | 12 | 12 | | 13 | 59 | | 14 | 2 | | 15 | 12 | | 16 | 25 | | 17 | 15 | | 18 | 5 | | 19 | 3 | | 20 | 50 | | 21 | 19 | | 22 | 2 | | 23 | 1 | | 24 | 28 | | 25 | 36 | | 26 | 2 | | 27 | 12 | | 28 | 28 | | 29 | 5 | | 30 | 10 | | 31 | 9 | | 32 | 4 | | 33 | 11 | | 34 | 7 | | 35 | 5 | | 36 | 2 | | 37 | 32 | | 38 | 21 | | 39 | 5 | | 40 | 2 | | 41 | 3 | | 42 | 85 | | 43 | 25 | | 44 | 9 | | 45 | 38 | | 46 | 4 | | 47 | 30 | | 48 | 9 | | 49 | 9 |
| |
| 86.55% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 4 | | totalSentences | 75 | | matches | | 0 | "were curled" | | 1 | "been cleared" | | 2 | "was rushed" | | 3 | "were rehearsed" | | 4 | "was carved" |
| |
| 5.35% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 4 | | totalVerbs | 137 | | matches | | 0 | "was doing" | | 1 | "were defending" | | 2 | "were getting" | | 3 | "was photographing" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 0 | | semicolonCount | 0 | | flaggedSentences | 0 | | totalSentences | 113 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 804 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 26 | | adverbRatio | 0.03233830845771144 | | lyAdverbCount | 8 | | lyAdverbRatio | 0.009950248756218905 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 113 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 113 | | mean | 11.29 | | std | 8.4 | | cv | 0.744 | | sampleLengths | | 0 | 22 | | 1 | 25 | | 2 | 20 | | 3 | 13 | | 4 | 13 | | 5 | 3 | | 6 | 20 | | 7 | 22 | | 8 | 32 | | 9 | 11 | | 10 | 3 | | 11 | 16 | | 12 | 3 | | 13 | 7 | | 14 | 15 | | 15 | 10 | | 16 | 9 | | 17 | 4 | | 18 | 11 | | 19 | 2 | | 20 | 16 | | 21 | 13 | | 22 | 3 | | 23 | 9 | | 24 | 12 | | 25 | 5 | | 26 | 12 | | 27 | 35 | | 28 | 6 | | 29 | 12 | | 30 | 14 | | 31 | 13 | | 32 | 12 | | 33 | 16 | | 34 | 4 | | 35 | 2 | | 36 | 12 | | 37 | 25 | | 38 | 3 | | 39 | 12 | | 40 | 5 | | 41 | 3 | | 42 | 31 | | 43 | 7 | | 44 | 12 | | 45 | 6 | | 46 | 13 | | 47 | 2 | | 48 | 1 | | 49 | 19 |
| |
| 84.37% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 3 | | diversityRatio | 0.5221238938053098 | | totalSentences | 113 | | uniqueOpeners | 59 | |
| 95.24% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 2 | | totalSentences | 70 | | matches | | 0 | "Too long for innocent confusion," | | 1 | "Then she felt the bone" |
| | ratio | 0.029 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 15 | | totalSentences | 70 | | matches | | 0 | "She'd heard rumours about this" | | 1 | "She pulled on latex gloves" | | 2 | "His fingers were curled around" | | 3 | "She photographed it in situ" | | 4 | "She walked the perimeter of" | | 5 | "She didn't touch it." | | 6 | "She kept tucking her hair" | | 7 | "She sat down on the" | | 8 | "She'd never trusted digital records" | | 9 | "She reached into her satchel" | | 10 | "It was carved with symbols" | | 11 | "She walked back to the" | | 12 | "She thought about DS Morris," | | 13 | "She held up the compass" | | 14 | "Her phone buzzed." |
| | ratio | 0.214 | |
| 38.57% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 59 | | totalSentences | 70 | | matches | | 0 | "The body lay spread-eagled across" | | 1 | "Detective Harlow Quinn ducked under" | | 2 | "DS Chen handed her a" | | 3 | "Quinn circled the body, her" | | 4 | "The Veil Market." | | 5 | "She'd heard rumours about this" | | 6 | "An underground market that moved" | | 7 | "Chen pointed to the wound" | | 8 | "Quinn crouched beside the body," | | 9 | "Someone had drawn these before" | | 10 | "Symbols she didn't recognise, angular" | | 11 | "She pulled on latex gloves" | | 12 | "His fingers were curled around" | | 13 | "Brass, from the colour." | | 14 | "She photographed it in situ" | | 15 | "The needle spun lazily, then" | | 16 | "Quinn slipped it into an" | | 17 | "Chen looked at her sideways." | | 18 | "Quinn stood, her knees protesting" | | 19 | "The evidence, unfortunately, was doing" |
| | ratio | 0.843 | |
| 100.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 3 | | totalSentences | 70 | | matches | | 0 | "Now here she stood, in" | | 1 | "Now they were getting somewhere." | | 2 | "If you want answers, come" |
| | ratio | 0.043 | |
| 50.69% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 31 | | technicalSentenceCount | 4 | | matches | | 0 | "An underground market that moved locations every full moon, dealing in things that didn't appear in any evidence catalogue she'd ever seen." | | 1 | "Now here she stood, in what should have been a derelict Tube station beneath Camden, surrounded by empty stalls draped in velvet and chains of brass lanterns th…" | | 2 | "Symbols she didn't recognise, angular things that seemed to shift when viewed from different angles." | | 3 | "A leather-bound book with pages that felt wrong, too thick, too warm, lay open on another." |
| |
| 89.29% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 14 | | uselessAdditionCount | 1 | | matches | | 0 | "Quinn stood, her knees protesting" |
| |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | |