| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 1 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 4 | | tagDensity | 0.25 | | leniency | 0.5 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 92.65% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1360 | | 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) | |
| 15.44% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1360 | | totalAiIsms | 23 | | found | | | highlights | | 0 | "pulse" | | 1 | "racing" | | 2 | "traced" | | 3 | "familiar" | | 4 | "perfect" | | 5 | "chill" | | 6 | "navigating" | | 7 | "pulsed" | | 8 | "silence" | | 9 | "whisper" | | 10 | "fractured" | | 11 | "measured" | | 12 | "oppressive" | | 13 | "weight" | | 14 | "trembled" |
| |
| 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 | 137 | | matches | (empty) | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 0 | | hedgeCount | 3 | | narrationSentences | 137 | | filterMatches | (empty) | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 140 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 31 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 0 | | markdownWords | 0 | | totalWords | 1351 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 5 | | unquotedAttributions | 0 | | matches | (empty) | |
| 89.98% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 44 | | wordCount | 1333 | | uniqueNames | 13 | | maxNameDensity | 1.2 | | worstName | "Rory" | | maxWindowNameDensity | 2 | | worstWindowName | "Eva" | | discoveredNames | | Heartstone | 3 | | Fae | 1 | | Grove | 1 | | Eva | 11 | | Silas | 2 | | November | 1 | | Rory | 16 | | Park | 2 | | London | 1 | | Cardiff | 2 | | Pre-Law | 1 | | Evan | 2 | | Aurora | 1 |
| | persons | | 0 | "Eva" | | 1 | "Silas" | | 2 | "November" | | 3 | "Rory" | | 4 | "Evan" |
| | places | | 0 | "Fae" | | 1 | "Grove" | | 2 | "Park" | | 3 | "London" | | 4 | "Cardiff" |
| | globalScore | 0.9 | | windowScore | 1 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 102 | | glossingSentenceCount | 0 | | matches | (empty) | |
| 51.96% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 2 | | per1kWords | 1.48 | | wordCount | 1351 | | matches | | 0 | "Not just quiet, but a deep, ringing absence of sound" | | 1 | "not with bruising force, but with an implacable, impersonal strength, like roots closing" |
| |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 1 | | totalSentences | 140 | | matches | | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 54 | | mean | 25.02 | | std | 21.9 | | cv | 0.875 | | sampleLengths | | 0 | 8 | | 1 | 62 | | 2 | 5 | | 3 | 85 | | 4 | 54 | | 5 | 7 | | 6 | 33 | | 7 | 70 | | 8 | 1 | | 9 | 81 | | 10 | 8 | | 11 | 10 | | 12 | 42 | | 13 | 1 | | 14 | 12 | | 15 | 48 | | 16 | 8 | | 17 | 45 | | 18 | 3 | | 19 | 35 | | 20 | 19 | | 21 | 27 | | 22 | 21 | | 23 | 4 | | 24 | 50 | | 25 | 54 | | 26 | 7 | | 27 | 37 | | 28 | 42 | | 29 | 8 | | 30 | 49 | | 31 | 1 | | 32 | 36 | | 33 | 6 | | 34 | 19 | | 35 | 1 | | 36 | 51 | | 37 | 20 | | 38 | 7 | | 39 | 43 | | 40 | 26 | | 41 | 4 | | 42 | 14 | | 43 | 12 | | 44 | 6 | | 45 | 36 | | 46 | 12 | | 47 | 49 | | 48 | 2 | | 49 | 14 |
| |
| 95.02% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 4 | | totalSentences | 137 | | matches | | 0 | "was supposed" | | 1 | "was followed" | | 2 | "was gone" | | 3 | "was buried" |
| |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 2 | | totalVerbs | 204 | | matches | | 0 | "was trying" | | 1 | "was leaving" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 0 | | semicolonCount | 0 | | flaggedSentences | 0 | | totalSentences | 140 | | ratio | 0 | | matches | (empty) | |
| 88.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1346 | | adjectiveStacks | 2 | | stackExamples | | 0 | "damp, spring-soft earth." | | 1 | "warm, alive under her" |
| | adverbCount | 46 | | adverbRatio | 0.03417533432392273 | | lyAdverbCount | 7 | | lyAdverbRatio | 0.005200594353640416 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 140 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 140 | | mean | 9.65 | | std | 6.16 | | cv | 0.638 | | sampleLengths | | 0 | 8 | | 1 | 19 | | 2 | 7 | | 3 | 18 | | 4 | 18 | | 5 | 5 | | 6 | 5 | | 7 | 24 | | 8 | 23 | | 9 | 17 | | 10 | 16 | | 11 | 14 | | 12 | 16 | | 13 | 24 | | 14 | 7 | | 15 | 17 | | 16 | 8 | | 17 | 8 | | 18 | 4 | | 19 | 15 | | 20 | 18 | | 21 | 24 | | 22 | 4 | | 23 | 5 | | 24 | 1 | | 25 | 8 | | 26 | 8 | | 27 | 5 | | 28 | 13 | | 29 | 7 | | 30 | 21 | | 31 | 6 | | 32 | 13 | | 33 | 8 | | 34 | 10 | | 35 | 2 | | 36 | 3 | | 37 | 6 | | 38 | 18 | | 39 | 13 | | 40 | 1 | | 41 | 9 | | 42 | 3 | | 43 | 11 | | 44 | 11 | | 45 | 15 | | 46 | 11 | | 47 | 8 | | 48 | 10 | | 49 | 6 |
| |
| 41.43% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 12 | | diversityRatio | 0.3 | | totalSentences | 140 | | uniqueOpeners | 42 | |
| 100.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 4 | | totalSentences | 124 | | matches | | 0 | "Just that perpetual, false dusk." | | 1 | "Then, a whisper." | | 2 | "Just a sigh of vowels" | | 3 | "Then the ground beneath her" |
| | ratio | 0.032 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 35 | | totalSentences | 124 | | matches | | 0 | "It lay warm against her" | | 1 | "She traced the crescent-shaped scar" | | 2 | "She was supposed to be" | | 3 | "His knuckles had been white" | | 4 | "She’d arrived with ten minutes" | | 5 | "She checked her phone for" | | 6 | "It had been 6:17 for" | | 7 | "Her delivery shift instincts screamed" | | 8 | "She was used to navigating" | | 9 | "It was followed by a" | | 10 | "Her voice sounded small, swallowed" | | 11 | "Her own breathing seemed too" | | 12 | "It came from everywhere and" | | 13 | "Her heel sank into the" | | 14 | "It had a cadence, a" | | 15 | "It was trying to form" | | 16 | "She turned, fully intending to" | | 17 | "It didn't budge." | | 18 | "She swallowed it down." | | 19 | "She forced herself to breathe," |
| | ratio | 0.282 | |
| 68.87% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 97 | | totalSentences | 124 | | matches | | 0 | "Rory's left hand closed around" | | 1 | "It lay warm against her" | | 2 | "The silver chain felt cold" | | 3 | "She traced the crescent-shaped scar" | | 4 | "The familiar texture grounded her," | | 5 | "The Fae Grove was wrong." | | 6 | "The ancient oak standing stones" | | 7 | "Frost crunched under Rory's boots" | | 8 | "The air smelled of crushed" | | 9 | "Richmond Park lay silent beyond" | | 10 | "The distant hum of London" | | 11 | "Here, there was only the" | | 12 | "She was supposed to be" | | 13 | "Silas had grunted, polishing a" | | 14 | "His knuckles had been white" | | 15 | "Rory hadn't been late." | | 16 | "She’d arrived with ten minutes" | | 17 | "The sky beyond the stone" | | 18 | "She checked her phone for" | | 19 | "The time displayed 6:17 PM." |
| | ratio | 0.782 | |
| 100.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 3 | | totalSentences | 124 | | matches | | 0 | "Even when she'd told him" | | 1 | "Even when Evan had shown" | | 2 | "If she kept the stones" |
| | ratio | 0.024 | |
| 72.60% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 61 | | technicalSentenceCount | 6 | | matches | | 0 | "It lay warm against her collarbone, a steady pulse against her skin that didn't match her own racing heart." | | 1 | "The ancient oak standing stones formed a perfect circle, guarding a clearing of wildflowers that bloomed in defiance of the late November chill." | | 2 | "Here, there was only the rustle of leaves that shouldn't exist on bare winter branches, and the occasional snap of a twick-twack stick underfoot." | | 3 | "It came from everywhere and nowhere, a breath of sound that skittered across the back of her neck." | | 4 | "Where two great standing stones had framed a perfect exit, a solid wall of ancient, gnarled oak now grew, bark fused together as if it had been that way for cen…" | | 5 | "Rory crashed through low-hanging branches that scraped at her arms, her breath coming in ragged gasps." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 1 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 0.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 1 | | fancyCount | 1 | | fancyTags | | | dialogueSentences | 4 | | tagDensity | 0.25 | | leniency | 0.5 | | rawRatio | 1 | | effectiveRatio | 0.5 | |