| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | |
| 71.35% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1745 | | totalAiIsmAdverbs | 10 | | found | | | highlights | | 0 | "tightly" | | 1 | "slightly" | | 2 | "nervously" | | 3 | "really" | | 4 | "carefully" | | 5 | "slowly" | | 6 | "hesitantly" | | 7 | "completely" | | 8 | "very" |
| |
| 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) | |
| 65.62% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1745 | | totalAiIsms | 12 | | found | | | highlights | | 0 | "sanctuary" | | 1 | "echoing" | | 2 | "measured" | | 3 | "cataloged" | | 4 | "chaotic" | | 5 | "firmly" | | 6 | "calculated" | | 7 | "etched" | | 8 | "intricate" | | 9 | "facade" | | 10 | "shattered" |
| |
| 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 | 145 | | matches | (empty) | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 3 | | hedgeCount | 1 | | narrationSentences | 145 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 145 | | 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 | 0 | | markdownWords | 0 | | totalWords | 1732 | | ratio | 0 | | matches | (empty) | |
| 0.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 9 | | unquotedAttributions | 9 | | matches | | 0 | "I know what this is, Eva said, her voice trembling slightly in the cavernous acoustic space." | | 1 | "It is a Veil Compass, Eva explained, shifting the heavy strap of her satchel." | | 2 | "But the compass, Eva insisted, stepping closer." | | 3 | "The casing has a patina of verdigris, she murmured." | | 4 | "Exactly, Harlow said." | | 5 | "Yes, Eva agreed hesitantly." | | 6 | "I suppose, Eva said, her voice pitching up in realization." | | 7 | "The needle never lies, Harlow said, her voice ringing out clearly over the hum of the portable halogen lights." | | 8 | "You and I are going to have a very long conversation about who is running these tokens in the clique, Harlow continued, …" |
| |
| 63.39% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 92 | | wordCount | 1732 | | uniqueNames | 24 | | maxNameDensity | 1.73 | | worstName | "Harlow" | | maxWindowNameDensity | 3 | | worstWindowName | "Eva" | | discoveredNames | | Tube | 3 | | Camden | 2 | | Harlow | 30 | | Quinn | 2 | | Veil | 4 | | Market | 2 | | Metropolitan | 1 | | Police | 1 | | London | 2 | | Shade | 2 | | Kowalski | 3 | | Eva | 22 | | Met | 1 | | Aurora | 3 | | Ancient | 1 | | History | 1 | | Oxford | 1 | | British | 1 | | Museum | 1 | | Detective | 3 | | Thames | 1 | | Compass | 2 | | Miss | 1 | | Morris | 2 |
| | persons | | 0 | "Camden" | | 1 | "Harlow" | | 2 | "Quinn" | | 3 | "Market" | | 4 | "Police" | | 5 | "Kowalski" | | 6 | "Eva" | | 7 | "Aurora" | | 8 | "Museum" | | 9 | "Detective" | | 10 | "Miss" | | 11 | "Morris" |
| | places | | 0 | "London" | | 1 | "Ancient" | | 2 | "Oxford" | | 3 | "British" | | 4 | "Thames" | | 5 | "Compass" |
| | globalScore | 0.634 | | windowScore | 0.667 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 117 | | glossingSentenceCount | 1 | | matches | | 0 | "sigils that seemed to catch and refract the halogen light" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 1732 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 145 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 39 | | mean | 44.41 | | std | 28.98 | | cv | 0.653 | | sampleLengths | | 0 | 98 | | 1 | 66 | | 2 | 65 | | 3 | 30 | | 4 | 45 | | 5 | 17 | | 6 | 77 | | 7 | 142 | | 8 | 23 | | 9 | 26 | | 10 | 73 | | 11 | 48 | | 12 | 16 | | 13 | 70 | | 14 | 36 | | 15 | 74 | | 16 | 32 | | 17 | 13 | | 18 | 92 | | 19 | 21 | | 20 | 43 | | 21 | 11 | | 22 | 66 | | 23 | 46 | | 24 | 16 | | 25 | 52 | | 26 | 16 | | 27 | 23 | | 28 | 71 | | 29 | 11 | | 30 | 31 | | 31 | 9 | | 32 | 16 | | 33 | 58 | | 34 | 21 | | 35 | 28 | | 36 | 64 | | 37 | 31 | | 38 | 55 |
| |
| 76.23% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 12 | | totalSentences | 145 | | matches | | 0 | "were bathed" | | 1 | "been gutted" | | 2 | "were involved" | | 3 | "was etched" | | 4 | "was locked" | | 5 | "was opened" | | 6 | "is scraped" | | 7 | "was dropped" | | 8 | "are undisturbed" | | 9 | "was placed" | | 10 | "is warded" | | 11 | "are carved" |
| |
| 100.00% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 2 | | totalVerbs | 295 | | matches | | 0 | "were looking" | | 1 | "was pointing" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 0 | | semicolonCount | 0 | | flaggedSentences | 0 | | totalSentences | 145 | | ratio | 0 | | matches | (empty) | |
| 97.56% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1745 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 61 | | adverbRatio | 0.03495702005730659 | | lyAdverbCount | 40 | | lyAdverbRatio | 0.022922636103151862 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 145 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 145 | | mean | 11.94 | | std | 6.79 | | cv | 0.568 | | sampleLengths | | 0 | 16 | | 1 | 35 | | 2 | 5 | | 3 | 15 | | 4 | 27 | | 5 | 10 | | 6 | 7 | | 7 | 16 | | 8 | 18 | | 9 | 15 | | 10 | 14 | | 11 | 23 | | 12 | 10 | | 13 | 18 | | 14 | 9 | | 15 | 17 | | 16 | 4 | | 17 | 29 | | 18 | 16 | | 19 | 5 | | 20 | 12 | | 21 | 11 | | 22 | 15 | | 23 | 18 | | 24 | 19 | | 25 | 14 | | 26 | 19 | | 27 | 12 | | 28 | 18 | | 29 | 23 | | 30 | 23 | | 31 | 5 | | 32 | 14 | | 33 | 28 | | 34 | 23 | | 35 | 16 | | 36 | 5 | | 37 | 5 | | 38 | 4 | | 39 | 11 | | 40 | 20 | | 41 | 17 | | 42 | 21 | | 43 | 14 | | 44 | 6 | | 45 | 18 | | 46 | 10 | | 47 | 4 | | 48 | 12 | | 49 | 9 |
| |
| 50.11% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 9 | | diversityRatio | 0.33793103448275863 | | totalSentences | 145 | | uniqueOpeners | 49 | |
| 100.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 5 | | totalSentences | 143 | | matches | | 0 | "Officially, the Metropolitan Police did" | | 1 | "Unofficially, Harlow had spent the" | | 2 | "Instead, the echoing tunnels were" | | 3 | "Exactly, Harlow said." | | 4 | "Just a desperate thief crawling" |
| | ratio | 0.035 | |
| 100.00% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 35 | | totalSentences | 143 | | matches | | 0 | "She stepped under the yellow" | | 1 | "He was a Shade artisan," | | 2 | "He had been gutted." | | 3 | "She was small, only five-foot-four," | | 4 | "Her green eyes were wide," | | 5 | "It was a tell Harlow" | | 6 | "She held a master degree" | | 7 | "I know what this is," | | 8 | "She pointed toward the floor." | | 9 | "Its face was etched with" | | 10 | "It is a Veil Compass," | | 11 | "It means the killer was" | | 12 | "it could have been the" | | 13 | "She would not let ghost" | | 14 | "Her military precision took over" | | 15 | "It indicates a weapon swung" | | 16 | "She did not touch it," | | 17 | "It scratched against the tiles." | | 18 | "It fell from a height." | | 19 | "They did not." |
| | ratio | 0.245 | |
| 100.00% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 101 | | totalSentences | 143 | | matches | | 0 | "The air in the abandoned" | | 1 | "Detective Harlow Quinn stood at" | | 2 | "This was the Veil Market." | | 3 | "Harlow adjusted the worn leather" | | 4 | "The hands read three in" | | 5 | "The market moved locations every" | | 6 | "The vast underground cavern should" | | 7 | "She stepped under the yellow" | | 8 | "A collapsed merchant stall dominated" | | 9 | "He was a Shade artisan," | | 10 | "He had been gutted." | | 11 | "Harlow stood over the body," | | 12 | "The evidence painted a chaotic" | | 13 | "A throat cleared behind her." | | 14 | "Harlow turned, her closely cropped" | | 15 | "Eva Kowalski stood just inside" | | 16 | "She was small, only five-foot-four," | | 17 | "Her green eyes were wide," | | 18 | "The harsh lighting washed out" | | 19 | "Harlow watched closely as Eva" |
| | ratio | 0.706 | |
| 100.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 6 | | totalSentences | 143 | | matches | | 0 | "If a monster came through" | | 1 | "If it was dropped during" | | 2 | "If you do not have" | | 3 | "If they opened a portal," | | 4 | "If the energy was dense" | | 5 | "Because there are no monsters" |
| | ratio | 0.042 | |
| 100.00% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 83 | | technicalSentenceCount | 3 | | matches | | 0 | "Splintered wooden crates spilled glowing, viscous liquids that hissed violently as they ate into the century-old subway tiles." | | 1 | "He was a Shade artisan, his skin carrying a translucent, violet hue that marked his supernatural lineage." | | 2 | "Eva was Aurora's childhood friend, an occult researcher who had moved to London two years before Aurora arrived." |
| |
| 100.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 0 | | uselessAdditionCount | 0 | | matches | (empty) | |
| 100.00% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | |