| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 24 | | adverbTagCount | 2 | | adverbTags | | 0 | "he said finally [finally]" | | 1 | "Silas said quietly [quietly]" |
| | dialogueSentences | 67 | | tagDensity | 0.358 | | leniency | 0.716 | | rawRatio | 0.083 | | effectiveRatio | 0.06 | |
| 96.23% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1328 | | totalAiIsmAdverbs | 1 | | found | | | highlights | | |
| 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) | |
| 58.58% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1328 | | totalAiIsms | 11 | | found | | | highlights | | 0 | "silence" | | 1 | "flicked" | | 2 | "flickered" | | 3 | "weight" | | 4 | "echoed" | | 5 | "traced" | | 6 | "shattered" | | 7 | "pulse" | | 8 | "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 | 95 | | matches | (empty) | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 0 | | narrationSentences | 95 | | filterMatches | (empty) | | hedgeMatches | (empty) | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 136 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 35 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 4 | | markdownWords | 21 | | totalWords | 1322 | | ratio | 0.016 | | matches | | 0 | "\"Use it if you need to disappear.\"" | | 1 | "\"Go.\"" | | 2 | "\"Local Barrister’s Son Arrested in Domestic Assault Case.\"" | | 3 | "\"Gone. Don’t look for me.\"" |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 21 | | unquotedAttributions | 0 | | matches | (empty) | |
| 77.61% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 43 | | wordCount | 967 | | uniqueNames | 11 | | maxNameDensity | 1.45 | | worstName | "Silas" | | maxWindowNameDensity | 2.5 | | worstWindowName | "Rory" | | discoveredNames | | Rory | 13 | | Golden | 1 | | Empress | 1 | | Three | 1 | | Cardiff | 2 | | Eva | 1 | | Raven | 2 | | Nest | 2 | | Silas | 14 | | London | 3 | | Evan | 3 |
| | persons | | 0 | "Rory" | | 1 | "Empress" | | 2 | "Eva" | | 3 | "Raven" | | 4 | "Nest" | | 5 | "Silas" | | 6 | "Evan" |
| | places | | 0 | "Golden" | | 1 | "Cardiff" | | 2 | "London" |
| | globalScore | 0.776 | | windowScore | 0.833 | |
| 38.06% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 67 | | glossingSentenceCount | 3 | | matches | | 0 | "looked like they’d died before they hit t" | | 1 | "quite hide the old scar near his ear" | | 2 | "felt like an anchor, and for the first" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 1322 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 1 | | totalSentences | 136 | | matches | | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 66 | | mean | 20.03 | | std | 16.51 | | cv | 0.824 | | sampleLengths | | 0 | 49 | | 1 | 1 | | 2 | 57 | | 3 | 62 | | 4 | 21 | | 5 | 15 | | 6 | 15 | | 7 | 46 | | 8 | 8 | | 9 | 15 | | 10 | 25 | | 11 | 4 | | 12 | 18 | | 13 | 13 | | 14 | 70 | | 15 | 6 | | 16 | 14 | | 17 | 7 | | 18 | 9 | | 19 | 21 | | 20 | 7 | | 21 | 26 | | 22 | 8 | | 23 | 6 | | 24 | 12 | | 25 | 14 | | 26 | 15 | | 27 | 29 | | 28 | 7 | | 29 | 6 | | 30 | 4 | | 31 | 9 | | 32 | 60 | | 33 | 18 | | 34 | 32 | | 35 | 6 | | 36 | 2 | | 37 | 41 | | 38 | 3 | | 39 | 19 | | 40 | 57 | | 41 | 8 | | 42 | 13 | | 43 | 20 | | 44 | 28 | | 45 | 9 | | 46 | 11 | | 47 | 23 | | 48 | 10 | | 49 | 42 |
| |
| 100.00% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 1 | | totalSentences | 95 | | matches | | |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 2 | | totalVerbs | 172 | | matches | | 0 | "was joking" | | 1 | "was, threading" |
| |
| 58.82% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 5 | | semicolonCount | 0 | | flaggedSentences | 4 | | totalSentences | 136 | | ratio | 0.029 | | matches | | 0 | "The signet ring on his right hand caught the dim light as he lifted a hand in greeting—or was it surrender?" | | 1 | "Then he slid onto the stool beside her, close enough that she caught the scent of bergamot and something sharper—gun oil, maybe, or the metallic tang of old secrets." | | 2 | "The Raven’s Nest hadn’t changed—same old maps on the walls, same black-and-white photos of men who looked like they’d died before they hit thirty." | | 3 | "She’d packed her bag while he was out, left a note on the pillow—*\"Gone. Don’t look for me.\"*—and slipped out the back door." |
| |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 972 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 33 | | adverbRatio | 0.033950617283950615 | | lyAdverbCount | 5 | | lyAdverbRatio | 0.0051440329218107 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 136 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 136 | | mean | 9.72 | | std | 6.92 | | cv | 0.712 | | sampleLengths | | 0 | 16 | | 1 | 10 | | 2 | 23 | | 3 | 1 | | 4 | 18 | | 5 | 21 | | 6 | 18 | | 7 | 11 | | 8 | 22 | | 9 | 13 | | 10 | 16 | | 11 | 21 | | 12 | 10 | | 13 | 5 | | 14 | 6 | | 15 | 9 | | 16 | 2 | | 17 | 15 | | 18 | 29 | | 19 | 8 | | 20 | 12 | | 21 | 3 | | 22 | 13 | | 23 | 9 | | 24 | 3 | | 25 | 4 | | 26 | 16 | | 27 | 2 | | 28 | 8 | | 29 | 5 | | 30 | 3 | | 31 | 6 | | 32 | 20 | | 33 | 24 | | 34 | 3 | | 35 | 14 | | 36 | 6 | | 37 | 7 | | 38 | 7 | | 39 | 7 | | 40 | 9 | | 41 | 15 | | 42 | 6 | | 43 | 7 | | 44 | 8 | | 45 | 6 | | 46 | 12 | | 47 | 8 | | 48 | 6 | | 49 | 6 |
| |
| 65.93% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 7 | | diversityRatio | 0.4264705882352941 | | totalSentences | 136 | | uniqueOpeners | 58 | |
| 100.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 3 | | totalSentences | 88 | | matches | | 0 | "Then he slid onto the" | | 1 | "Instead, he reached into his" | | 2 | "Then his hand closed over" |
| | ratio | 0.034 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 26 | | totalSentences | 88 | | matches | | 0 | "His beard had more grey" | | 1 | "She’d laughed, thinking he was" | | 2 | "he said, voice rough as" | | 3 | "She forced a smile" | | 4 | "She signalled the bartender for" | | 5 | "He tapped the bar, two" | | 6 | "His gaze flicked to the" | | 7 | "She swirled her drink" | | 8 | "He laughed, but it was" | | 9 | "he said finally" | | 10 | "He rubbed his temple, the" | | 11 | "She leaned in, close enough" | | 12 | "She should’ve left it there." | | 13 | "she said, forcing her voice" | | 14 | "He slid it across the" | | 15 | "She unfolded it carefully." | | 16 | "Her breath hitched." | | 17 | "She’d packed her bag while" | | 18 | "She’d always wondered if he’d" | | 19 | "she said, voice steadier than" |
| | ratio | 0.295 | |
| 45.23% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 73 | | totalSentences | 88 | | matches | | 0 | "The glass slipped from Rory’s" | | 1 | "The ice cubes clinked against" | | 2 | "His beard had more grey" | | 3 | "The signet ring on his" | | 4 | "The limp in his left" | | 5 | "Rory wiped her hand on" | | 6 | "The last time she’d seen" | | 7 | "She’d laughed, thinking he was" | | 8 | "he said, voice rough as" | | 9 | "She forced a smile" | | 10 | "The air between them thickened," | | 11 | "She signalled the bartender for" | | 12 | "He tapped the bar, two" | | 13 | "The bartender nodded, already reaching" | | 14 | "His gaze flicked to the" | | 15 | "Rory flexed her fingers around" | | 16 | "The whisky arrived." | | 17 | "Silas took it neat, no" | | 18 | "Rory watched the way his" | | 19 | "The Raven’s Nest hadn’t changed—same" |
| | ratio | 0.83 | |
| 100.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 2 | | totalSentences | 88 | | matches | | 0 | "Now here he was, threading" | | 1 | "By the time he’d come" |
| | ratio | 0.023 | |
| 100.00% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 35 | | technicalSentenceCount | 1 | | matches | | 0 | "The Raven’s Nest hadn’t changed—same old maps on the walls, same black-and-white photos of men who looked like they’d died before they hit thirty." |
| |
| 62.50% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 24 | | uselessAdditionCount | 3 | | matches | | 0 | "he said, voice rough as gravel under tyres" | | 1 | "He rubbed, the signet ring catching the light again" | | 2 | "she said, voice steadier than she felt" |
| |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 12 | | fancyCount | 3 | | fancyTags | | 0 | "She’d (would)" | | 1 | "Rory laughed (laugh)" | | 2 | "He pressed (press)" |
| | dialogueSentences | 67 | | tagDensity | 0.179 | | leniency | 0.358 | | rawRatio | 0.25 | | effectiveRatio | 0.09 | |