| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 18 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 37 | | tagDensity | 0.486 | | leniency | 0.973 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 95.93% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1229 | | totalAiIsmAdverbs | 1 | | 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) | |
| 51.18% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1229 | | totalAiIsms | 12 | | found | | | highlights | | 0 | "echoed" | | 1 | "oppressive" | | 2 | "stark" | | 3 | "standard" | | 4 | "crystalline" | | 5 | "scanned" | | 6 | "racing" | | 7 | "weight" | | 8 | "traced" | | 9 | "silence" |
| |
| 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 | 109 | | matches | (empty) | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 1 | | narrationSentences | 109 | | filterMatches | (empty) | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 126 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 45 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 2 | | markdownWords | 7 | | totalWords | 1218 | | ratio | 0.006 | | matches | | 0 | "Aethelred" | | 1 | "Ask your researcher about the Compass." |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 15 | | unquotedAttributions | 0 | | matches | (empty) | |
| 47.84% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 47 | | wordCount | 881 | | uniqueNames | 9 | | maxNameDensity | 2.04 | | worstName | "Harlow" | | maxWindowNameDensity | 3 | | worstWindowName | "Harlow" | | discoveredNames | | Harlow | 18 | | Quinn | 1 | | Camden | 1 | | Tube | 1 | | Davies | 11 | | Eva | 11 | | Kowalski | 2 | | Morris | 1 | | Rolex | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Camden" | | 3 | "Davies" | | 4 | "Eva" | | 5 | "Kowalski" | | 6 | "Morris" | | 7 | "Rolex" |
| | places | (empty) | | globalScore | 0.478 | | windowScore | 0.667 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 68 | | glossingSentenceCount | 0 | | matches | (empty) | |
| 35.80% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 2 | | per1kWords | 1.642 | | wordCount | 1218 | | matches | | 0 | "not towards the boarded-up exit, but to a section of the tiled wall" | | 1 | "not on Davies, but on Eva Kowalski" |
| |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 126 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 38 | | mean | 32.05 | | std | 21.43 | | cv | 0.669 | | sampleLengths | | 0 | 49 | | 1 | 78 | | 2 | 15 | | 3 | 52 | | 4 | 13 | | 5 | 27 | | 6 | 28 | | 7 | 74 | | 8 | 30 | | 9 | 17 | | 10 | 35 | | 11 | 84 | | 12 | 13 | | 13 | 37 | | 14 | 47 | | 15 | 48 | | 16 | 24 | | 17 | 61 | | 18 | 21 | | 19 | 45 | | 20 | 10 | | 21 | 3 | | 22 | 32 | | 23 | 43 | | 24 | 4 | | 25 | 6 | | 26 | 47 | | 27 | 28 | | 28 | 21 | | 29 | 10 | | 30 | 16 | | 31 | 68 | | 32 | 19 | | 33 | 7 | | 34 | 17 | | 35 | 39 | | 36 | 44 | | 37 | 6 |
| |
| 98.83% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 2 | | totalSentences | 109 | | matches | | 0 | "was crouched" | | 1 | "was broken" |
| |
| 19.82% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 4 | | totalVerbs | 148 | | matches | | 0 | "was working" | | 1 | "were treating" | | 2 | "was racing" | | 3 | "was piecing" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 0 | | semicolonCount | 0 | | flaggedSentences | 0 | | totalSentences | 126 | | ratio | 0 | | matches | (empty) | |
| 94.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 659 | | adjectiveStacks | 1 | | stackExamples | | 0 | "young under harsh lights," |
| | adverbCount | 15 | | adverbRatio | 0.02276176024279211 | | lyAdverbCount | 3 | | lyAdverbRatio | 0.004552352048558422 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 126 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 126 | | mean | 9.67 | | std | 7.3 | | cv | 0.755 | | sampleLengths | | 0 | 20 | | 1 | 6 | | 2 | 21 | | 3 | 2 | | 4 | 19 | | 5 | 17 | | 6 | 8 | | 7 | 18 | | 8 | 16 | | 9 | 10 | | 10 | 5 | | 11 | 6 | | 12 | 17 | | 13 | 29 | | 14 | 11 | | 15 | 2 | | 16 | 10 | | 17 | 17 | | 18 | 14 | | 19 | 14 | | 20 | 7 | | 21 | 9 | | 22 | 5 | | 23 | 12 | | 24 | 3 | | 25 | 1 | | 26 | 4 | | 27 | 11 | | 28 | 2 | | 29 | 1 | | 30 | 11 | | 31 | 8 | | 32 | 11 | | 33 | 6 | | 34 | 13 | | 35 | 4 | | 36 | 6 | | 37 | 7 | | 38 | 11 | | 39 | 24 | | 40 | 5 | | 41 | 16 | | 42 | 13 | | 43 | 16 | | 44 | 33 | | 45 | 1 | | 46 | 6 | | 47 | 7 | | 48 | 16 | | 49 | 21 |
| |
| 59.26% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 12 | | diversityRatio | 0.4126984126984127 | | totalSentences | 126 | | uniqueOpeners | 52 | |
| 35.09% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 1 | | totalSentences | 95 | | matches | | 0 | "Almost crystalline, with a faint," |
| | ratio | 0.011 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 28 | | totalSentences | 95 | | matches | | 0 | "It was wrong for the" | | 1 | "Her heavy-soled shoes echoed on" | | 2 | "His suit was expensive, charcoal" | | 3 | "She kept her eyes on" | | 4 | "She knelt, the rough concrete" | | 5 | "She didn’t wait for an" | | 6 | "She shot a look at" | | 7 | "They were treating it like" | | 8 | "She kept her gaze on" | | 9 | "She’d buried it, filed it" | | 10 | "She walked away from the" | | 11 | "She ran a gloved hand" | | 12 | "They were cold." | | 13 | "It was warm." | | 14 | "She looked back at the" | | 15 | "They didn’t suggest a struggle." | | 16 | "They suggested a procession." | | 17 | "she asked Davies without turning" | | 18 | "She turned her attention back" | | 19 | "Her eyes, a deep, assessing" |
| | ratio | 0.295 | |
| 38.95% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 80 | | totalSentences | 95 | | matches | | 0 | "The scent hit Harlow Quinn" | | 1 | "It was wrong for the" | | 2 | "Her heavy-soled shoes echoed on" | | 3 | "The place was a tomb," | | 4 | "A man lay sprawled on" | | 5 | "His suit was expensive, charcoal" | | 6 | "A junior officer, pale and" | | 7 | "Harlow’s gaze shifted beyond the" | | 8 | "DS Davies was crouched a" | | 9 | "Davies straightened up, his face" | | 10 | "Harlow’s voice was low, clipped" | | 11 | "She kept her eyes on" | | 12 | "Davies gestured to the body" | | 13 | "Harlow moved closer, her movements" | | 14 | "She knelt, the rough concrete" | | 15 | "The victim was middle-aged, well-groomed." | | 16 | "The stab wound was a" | | 17 | "The edges of the wound" | | 18 | "The cloying sweetness seemed to" | | 19 | "Eva said, her voice soft" |
| | ratio | 0.842 | |
| 0.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 0 | | totalSentences | 95 | | matches | (empty) | | ratio | 0 | |
| 23.81% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 30 | | technicalSentenceCount | 5 | | matches | | 0 | "The scent hit Harlow Quinn first, a cloying sweetness that clung to the back of her throat like cheap perfume." | | 1 | "She kept her eyes on Eva, who tucked a stray curl of red hair behind her ear." | | 2 | "Almost crystalline, with a faint, shimmering residue that caught the light." | | 3 | "Scuff marks on the floor led not towards the boarded-up exit, but to a section of the tiled wall that seemed darker, the grout between the tiles forming pattern…" | | 4 | "She was piecing it together, the military precision of her mind assembling fragments." |
| |
| 0.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 18 | | uselessAdditionCount | 6 | | matches | | 0 | "Davies straightened up, his face a mask of professional neutrality" | | 1 | "Eva said, her voice soft but clear" | | 2 | "Eva explained, her green eyes earnest behind her round glasses" | | 3 | "Eva said, a flash of defiance in her voice" | | 4 | "She took, her presence suddenly filling the space" | | 5 | "Harlow said, the truth settling on her with a cold, heavy certainty" |
| |
| 41.89% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 11 | | fancyCount | 4 | | fancyTags | | 0 | "Davies sighed (sigh)" | | 1 | "Harlow pressed (press)" | | 2 | "Eva explained (explain)" | | 3 | "she whispered (whisper)" |
| | dialogueSentences | 37 | | tagDensity | 0.297 | | leniency | 0.595 | | rawRatio | 0.364 | | effectiveRatio | 0.216 | |