| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 3 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 29 | | tagDensity | 0.103 | | leniency | 0.207 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 100.00% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1287 | | totalAiIsmAdverbs | 0 | | found | (empty) | | highlights | (empty) | |
| 80.00% | 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) | |
| 68.92% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1287 | | totalAiIsms | 8 | | found | | | highlights | | 0 | "methodical" | | 1 | "silence" | | 2 | "etched" | | 3 | "complex" | | 4 | "pulse" | | 5 | "echoed" |
| |
| 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 | 88 | | matches | (empty) | |
| 77.92% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 2 | | hedgeCount | 2 | | narrationSentences | 88 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 114 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 69 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 0 | | markdownWords | 0 | | totalWords | 1280 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 8 | | unquotedAttributions | 0 | | matches | (empty) | |
| 66.67% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 37 | | wordCount | 791 | | uniqueNames | 14 | | maxNameDensity | 1.64 | | worstName | "Quinn" | | maxWindowNameDensity | 3 | | worstWindowName | "Quinn" | | discoveredNames | | Harlow | 1 | | Quinn | 13 | | Tube | 2 | | Victorian | 1 | | Camden | 1 | | Sergeant | 1 | | Marcus | 1 | | Webb | 9 | | London | 1 | | Morris | 3 | | Crown | 1 | | Prosecution | 1 | | Service | 1 | | Afghanistan | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Tube" | | 3 | "Sergeant" | | 4 | "Marcus" | | 5 | "Webb" | | 6 | "Morris" |
| | places | | | globalScore | 0.678 | | windowScore | 0.667 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 58 | | 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 | 1280 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 114 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 49 | | mean | 26.12 | | std | 19.03 | | cv | 0.729 | | sampleLengths | | 0 | 7 | | 1 | 52 | | 2 | 62 | | 3 | 24 | | 4 | 8 | | 5 | 41 | | 6 | 5 | | 7 | 32 | | 8 | 68 | | 9 | 48 | | 10 | 9 | | 11 | 40 | | 12 | 59 | | 13 | 11 | | 14 | 27 | | 15 | 44 | | 16 | 43 | | 17 | 3 | | 18 | 7 | | 19 | 6 | | 20 | 57 | | 21 | 4 | | 22 | 19 | | 23 | 37 | | 24 | 8 | | 25 | 13 | | 26 | 39 | | 27 | 7 | | 28 | 40 | | 29 | 41 | | 30 | 7 | | 31 | 47 | | 32 | 3 | | 33 | 36 | | 34 | 10 | | 35 | 40 | | 36 | 2 | | 37 | 28 | | 38 | 42 | | 39 | 32 | | 40 | 19 | | 41 | 2 | | 42 | 48 | | 43 | 36 | | 44 | 6 | | 45 | 3 | | 46 | 26 | | 47 | 28 | | 48 | 4 |
| |
| 97.29% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 2 | | totalSentences | 88 | | matches | | 0 | "was tired" | | 1 | "was etched" |
| |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 2 | | totalVerbs | 139 | | matches | | 0 | "were doing" | | 1 | "was pulling" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 0 | | semicolonCount | 0 | | flaggedSentences | 0 | | totalSentences | 114 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 794 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 20 | | adverbRatio | 0.02518891687657431 | | lyAdverbCount | 5 | | lyAdverbRatio | 0.006297229219143577 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 114 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 114 | | mean | 11.23 | | std | 10.78 | | cv | 0.96 | | sampleLengths | | 0 | 7 | | 1 | 19 | | 2 | 1 | | 3 | 2 | | 4 | 14 | | 5 | 16 | | 6 | 6 | | 7 | 1 | | 8 | 27 | | 9 | 12 | | 10 | 16 | | 11 | 24 | | 12 | 8 | | 13 | 3 | | 14 | 26 | | 15 | 12 | | 16 | 5 | | 17 | 3 | | 18 | 20 | | 19 | 9 | | 20 | 68 | | 21 | 5 | | 22 | 12 | | 23 | 2 | | 24 | 11 | | 25 | 15 | | 26 | 3 | | 27 | 9 | | 28 | 40 | | 29 | 2 | | 30 | 15 | | 31 | 17 | | 32 | 9 | | 33 | 3 | | 34 | 13 | | 35 | 11 | | 36 | 13 | | 37 | 6 | | 38 | 8 | | 39 | 44 | | 40 | 6 | | 41 | 17 | | 42 | 10 | | 43 | 10 | | 44 | 3 | | 45 | 7 | | 46 | 6 | | 47 | 6 | | 48 | 6 | | 49 | 12 |
| |
| 73.45% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 7 | | diversityRatio | 0.4778761061946903 | | totalSentences | 113 | | uniqueOpeners | 54 | |
| 0.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 0 | | totalSentences | 73 | | matches | (empty) | | ratio | 0 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 17 | | totalSentences | 73 | | matches | | 0 | "She checked her worn leather" | | 1 | "Her nose caught copper and" | | 2 | "She didn't move." | | 3 | "It made him useful for" | | 4 | "He was tired, the kind" | | 5 | "His shirt bore coffee stains" | | 6 | "She noticed the victim's hands" | | 7 | "He hadn't fought." | | 8 | "His face remained composed, eyes" | | 9 | "He held up the evidence" | | 10 | "Its face was etched with" | | 11 | "She'd seen similar marks during" | | 12 | "His coffee-stained shirt shifted as" | | 13 | "He'd quit three years ago." | | 14 | "She'd learned not to trust" | | 15 | "He pulled out his phone" | | 16 | "She'd learned years ago that" |
| | ratio | 0.233 | |
| 1.10% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 67 | | totalSentences | 73 | | matches | | 0 | "The corpse wasn't supposed to" | | 1 | "Detective Harlow Quinn stood in" | | 2 | "Throat opened with surgical precision," | | 3 | "The bones of the crime" | | 4 | "She checked her worn leather" | | 5 | "The platform at Camden lay" | | 6 | "The body had been here" | | 7 | "Her nose caught copper and" | | 8 | "Detective Sergeant Marcus Webb emerged" | | 9 | "She didn't move." | | 10 | "Webb was good at his" | | 11 | "It made him useful for" | | 12 | "Webb's jaw tightened." | | 13 | "He was tired, the kind" | | 14 | "His shirt bore coffee stains" | | 15 | "Quinn crouched beside the body." | | 16 | "The incision across the throat" | | 17 | "The killer had known what" | | 18 | "She noticed the victim's hands" | | 19 | "He hadn't fought." |
| | ratio | 0.918 | |
| 68.49% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 1 | | totalSentences | 73 | | matches | | 0 | "Even now, eighty years later," |
| | ratio | 0.014 | |
| 63.49% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 36 | | technicalSentenceCount | 4 | | matches | | 0 | "Detective Sergeant Marcus Webb emerged from behind the forensic tent, holding a evidence bag between his thumb and forefinger as if it might bite." | | 1 | "Webb was good at his job, methodical in a way that bordered on obsessive, but he had the habit of presenting conclusions first and evidence second." | | 2 | "He was tired, the kind of tired that came from working backwards through a night that had no right answers." | | 3 | "Its face was etched with symbols that made Quinn's neck tighten." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 3 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 1 | | fancyCount | 0 | | fancyTags | (empty) | | dialogueSentences | 29 | | tagDensity | 0.034 | | leniency | 0.069 | | rawRatio | 0 | | effectiveRatio | 0 | |