| 100.00% | Adverbs in dialogue tags | Target: ≤10% dialogue tags with adverbs | | totalTags | 11 | | adverbTagCount | 0 | | adverbTags | (empty) | | dialogueSentences | 23 | | tagDensity | 0.478 | | leniency | 0.957 | | rawRatio | 0 | | effectiveRatio | 0 | |
| 88.99% | AI-ism adverb frequency | Target: <2% AI-ism adverbs (58 tracked) | | wordCount | 1362 | | totalAiIsmAdverbs | 3 | | 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) | |
| 41.26% | AI-ism word frequency | Target: <2% AI-ism words (290 tracked) | | wordCount | 1362 | | totalAiIsms | 16 | | found | | | highlights | | 0 | "echoed" | | 1 | "oppressive" | | 2 | "stark" | | 3 | "familiar" | | 4 | "pulse" | | 5 | "efficient" | | 6 | "weight" | | 7 | "intricate" | | 8 | "raced" | | 9 | "perfect" | | 10 | "whisper" | | 11 | "resolve" | | 12 | "etched" | | 13 | "tapestry" | | 14 | "scanning" | | 15 | "depths" |
| |
| 100.00% | Cliché density | Target: ≤1 cliche(s) per 800-word window | | totalCliches | 1 | | maxInWindow | 1 | | found | | 0 | | label | "blood ran cold" | | count | 1 |
|
| | highlights | | |
| 100.00% | Emotion telling (show vs. tell) | Target: ≤3% sentences with emotion telling | | emotionTells | 1 | | narrationSentences | 99 | | matches | | |
| 100.00% | Filter word density | Target: ≤3% sentences with filter/hedge words | | filterCount | 1 | | hedgeCount | 1 | | narrationSentences | 99 | | filterMatches | | | hedgeMatches | | |
| 100.00% | Gibberish response detection | Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words) | | analyzedSentences | 112 | | gibberishSentences | 0 | | adjustedGibberishSentences | 0 | | longSentenceCount | 0 | | runOnParagraphCount | 0 | | giantParagraphCount | 0 | | wordSaladCount | 0 | | repetitionLoopCount | 0 | | controlTokenCount | 0 | | maxSentenceWordsSeen | 33 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Markdown formatting overuse | Target: ≤5% words in markdown formatting | | markdownSpans | 3 | | markdownWords | 10 | | totalWords | 1358 | | ratio | 0.007 | | matches | | 0 | "Pointing" | | 1 | "them" | | 2 | "This is what happens to those who talk." |
| |
| 100.00% | Missing dialogue indicators (quotation marks) | Target: ≤10% speech attributions without quotation marks | | totalAttributions | 17 | | unquotedAttributions | 0 | | matches | (empty) | |
| 50.00% | Name drop frequency | Target: ≤1.0 per-name mentions per 100 words | | totalMentions | 38 | | wordCount | 1069 | | uniqueNames | 10 | | maxNameDensity | 1.5 | | worstName | "Quinn" | | maxWindowNameDensity | 3.5 | | worstWindowName | "Quinn" | | discoveredNames | | Camden | 1 | | Tube | 1 | | Harlow | 1 | | Quinn | 16 | | Miller | 7 | | Morris | 2 | | Kowalski | 1 | | Eva | 7 | | Alistair | 1 | | Finch | 1 |
| | persons | | 0 | "Harlow" | | 1 | "Quinn" | | 2 | "Miller" | | 3 | "Morris" | | 4 | "Kowalski" | | 5 | "Eva" | | 6 | "Alistair" | | 7 | "Finch" |
| | places | (empty) | | globalScore | 0.752 | | windowScore | 0.5 | |
| 100.00% | Narrator intent-glossing | Target: ≤2% narration sentences with intent-glossing patterns | | analyzedSentences | 78 | | glossingSentenceCount | 1 | | matches | | 0 | "looked like a child who had seen a monste" |
| |
| 100.00% | "Not X but Y" pattern overuse | Target: ≤1 "not X but Y" per 1000 words | | totalMatches | 0 | | per1kWords | 0 | | wordCount | 1358 | | matches | (empty) | |
| 100.00% | Overuse of "that" (subordinate clause padding) | Target: ≤2% sentences with "that" clauses | | thatCount | 0 | | totalSentences | 112 | | matches | (empty) | |
| 100.00% | Paragraph length variance | Target: CV ≥0.5 for paragraph word counts | | totalParagraphs | 30 | | mean | 45.27 | | std | 22.98 | | cv | 0.508 | | sampleLengths | | 0 | 90 | | 1 | 63 | | 2 | 62 | | 3 | 5 | | 4 | 44 | | 5 | 26 | | 6 | 22 | | 7 | 47 | | 8 | 36 | | 9 | 48 | | 10 | 61 | | 11 | 66 | | 12 | 40 | | 13 | 72 | | 14 | 19 | | 15 | 53 | | 16 | 26 | | 17 | 38 | | 18 | 41 | | 19 | 31 | | 20 | 33 | | 21 | 76 | | 22 | 29 | | 23 | 74 | | 24 | 18 | | 25 | 41 | | 26 | 55 | | 27 | 35 | | 28 | 101 | | 29 | 6 |
| |
| 91.09% | Passive voice overuse | Target: ≤2% passive sentences | | passiveCount | 4 | | totalSentences | 99 | | matches | | 0 | "was bunched" | | 1 | "was surprised" | | 2 | "been dismissed" | | 3 | "being fought" | | 4 | "been fired" |
| |
| 12.73% | Past progressive (was/were + -ing) overuse | Target: ≤2% past progressive verbs | | pastProgressiveCount | 5 | | totalVerbs | 178 | | matches | | 0 | "was beginning" | | 1 | "was stumbling" | | 2 | "was staring" | | 3 | "was seeing" | | 4 | "was starting" |
| |
| 100.00% | Em-dash & semicolon overuse | Target: ≤2% sentences with em-dashes/semicolons | | emDashCount | 0 | | semicolonCount | 0 | | flaggedSentences | 0 | | totalSentences | 112 | | ratio | 0 | | matches | (empty) | |
| 100.00% | Purple prose (modifier overload) | Target: <4% adverbs, <2% -ly adverbs, no adj stacking | | wordCount | 1094 | | adjectiveStacks | 0 | | stackExamples | (empty) | | adverbCount | 33 | | adverbRatio | 0.03016453382084095 | | lyAdverbCount | 10 | | lyAdverbRatio | 0.009140767824497258 | |
| 100.00% | Repeated phrase echo | Target: ≤20% sentences with echoes (window: 2) | | totalSentences | 112 | | echoCount | 0 | | echoWords | (empty) | |
| 100.00% | Sentence length variance | Target: CV ≥0.4 for sentence word counts | | totalSentences | 112 | | mean | 12.13 | | std | 7.65 | | cv | 0.631 | | sampleLengths | | 0 | 33 | | 1 | 25 | | 2 | 32 | | 3 | 17 | | 4 | 12 | | 5 | 25 | | 6 | 9 | | 7 | 9 | | 8 | 8 | | 9 | 9 | | 10 | 20 | | 11 | 16 | | 12 | 5 | | 13 | 21 | | 14 | 14 | | 15 | 9 | | 16 | 16 | | 17 | 6 | | 18 | 4 | | 19 | 22 | | 20 | 4 | | 21 | 17 | | 22 | 7 | | 23 | 19 | | 24 | 7 | | 25 | 29 | | 26 | 22 | | 27 | 10 | | 28 | 16 | | 29 | 6 | | 30 | 20 | | 31 | 7 | | 32 | 9 | | 33 | 19 | | 34 | 6 | | 35 | 8 | | 36 | 8 | | 37 | 8 | | 38 | 2 | | 39 | 3 | | 40 | 6 | | 41 | 3 | | 42 | 9 | | 43 | 13 | | 44 | 13 | | 45 | 5 | | 46 | 22 | | 47 | 25 | | 48 | 12 | | 49 | 22 |
| |
| 41.07% | Sentence opener variety | Target: ≥60% unique sentence openers | | consecutiveRepeats | 14 | | diversityRatio | 0.32142857142857145 | | totalSentences | 112 | | uniqueOpeners | 36 | |
| 0.00% | Adverb-first sentence starts | Target: ≥3% sentences starting with an adverb | | adverbCount | 0 | | totalSentences | 90 | | matches | (empty) | | ratio | 0 | |
| 42.22% | Pronoun-first sentence starts | Target: ≤30% sentences starting with a pronoun | | pronounCount | 40 | | totalSentences | 90 | | matches | | 0 | "Her polished boots echoed with" | | 1 | "It hadn’t mentioned the two" | | 2 | "It hadn’t mentioned the utter" | | 3 | "She didn’t need to check" | | 4 | "She pulled a pair of" | | 5 | "Her worn leather watch ticked" | | 6 | "He stepped into the light," | | 7 | "He’d been on the scene" | | 8 | "Her eyes never left the" | | 9 | "She leaned in closer, ignoring" | | 10 | "she repeated, her tone flat" | | 11 | "He’d been her partner for" | | 12 | "He didn’t say Morris’s name," | | 13 | "His ghost was always there," | | 14 | "She noted the way the" | | 15 | "His hands were clean, the" | | 16 | "She shifted her weight, and" | | 17 | "It was a small, circular" | | 18 | "She picked it up, brushing" | | 19 | "It felt unnervingly warm in" |
| | ratio | 0.444 | |
| 0.00% | Subject-first sentence starts | Target: ≤72% sentences starting with a subject | | subjectCount | 84 | | totalSentences | 90 | | matches | | 0 | "The air in the abandoned" | | 1 | "Her polished boots echoed with" | | 2 | "The crime scene techs had" | | 3 | "The body lay near the" | | 4 | "The initial call had been" | | 5 | "It hadn’t mentioned the two" | | 6 | "It hadn’t mentioned the utter" | | 7 | "Quinn crouched, her knees protesting" | | 8 | "She didn’t need to check" | | 9 | "The unnatural stillness of the" | | 10 | "She pulled a pair of" | | 11 | "Her worn leather watch ticked" | | 12 | "The voice came from the" | | 13 | "He stepped into the light," | | 14 | "He’d been on the scene" | | 15 | "Quinn replied, her voice low" | | 16 | "Her eyes never left the" | | 17 | "Miller said, gesturing vaguely with" | | 18 | "Quinn’s sharp jaw tightened." | | 19 | "She leaned in closer, ignoring" |
| | ratio | 0.933 | |
| 100.00% | Subordinate conjunction sentence starts | Target: ≥2% sentences starting with a subordinating conjunction | | subConjCount | 3 | | totalSentences | 90 | | matches | | 0 | "Before Miller could offer another" | | 1 | "To the people who knew" | | 2 | "Now, she was starting to" |
| | ratio | 0.033 | |
| 57.82% | Technical jargon density | Target: ≤6% sentences with technical-jargon patterns | | analyzedSentences | 42 | | technicalSentenceCount | 5 | | matches | | 0 | "The air in the abandoned Camden Tube station was a thick, cold soup that tasted of wet stone, rust, and something else, something metallic and coppery that Harl…" | | 1 | "The crime scene techs had set up their halogen lamps, and the stark white light they cast felt blasphemous here, illuminating a world that was never meant to be…" | | 2 | "The voice came from the shadows near an old ticket booth, layered with a wry weariness that belonged to DS Miller." | | 3 | "He’d been her partner for a year, a solid, by-the-book copper who thought her instincts were a relic of a traumatic past." | | 4 | "Before Miller could offer another dismissive retort, a new voice, freckled complexion pale with shock, and large, round glasses that magnified her wide, green e…" |
| |
| 0.00% | Useless dialogue additions | Target: ≤5% dialogue tags with trailing filler fragments | | totalTags | 11 | | uselessAdditionCount | 6 | | matches | | 0 | "Quinn replied, her voice low and even" | | 1 | "she repeated, her tone flat" | | 2 | "she said, her voice dropping to a near-whisper" | | 3 | "Quinn stood, her bearing straightening with military precision" | | 4 | "she said, her voice gaining a sliver of strength" | | 5 | "Quinn asked, her voice barely a breath" |
| |
| 19.57% | Dialogue tag variety (said vs. fancy) | Target: ≤10% fancy dialogue tags | | totalTags | 8 | | fancyCount | 3 | | fancyTags | | 0 | "she repeated (repeat)" | | 1 | "Eva stammered (stammer)" | | 2 | "she whispered (whisper)" |
| | dialogueSentences | 23 | | tagDensity | 0.348 | | leniency | 0.696 | | rawRatio | 0.375 | | effectiveRatio | 0.261 | |