Run: 2026-02-17-01KHMXPB470XD3XGPJG69KD4GR

Run Details

Model:: Claude 3.5 Haiku
Test:: Bad Writing Habits
Scenario:: Mystery: examining a crime scene
Duration:: 12.7s
Cost:: $0.002778

Model input

System

You are an expert novelist with a gift for vivid, immersive prose. You write in tight third-person limited point of view with a strong narrative voice. Your prose is clean, precise, and engaging — you favor strong verbs, concrete details, and natural-sounding dialogue.

You will be given a story bible (codex) containing canonical details about characters, locations, items, and lore. Use these details to ground your writing in the established world. Stay consistent with the codex.

Write only the prose of the scene. Do not include titles, chapter headings, markdown formatting, or any commentary before or after the scene.

User/Human

<codex>
<character id="4416" name="Detective Harlow Quinn" aliases="Quinn, Detective Quinn, Harlow" pronouns="she/her" species="Human" storyRole="Antagonist">
  <eyeColor>brown</eyeColor>
  <hair>closely cropped salt-and-pepper</hair>
  <height>5'9"</height>
  <age>41</age>
  <jawline>sharp jaw</jawline>
  <watch>worn leather watch on her left wrist</watch>
  <bearing>military precision</bearing>
  <description>A Metropolitan Police detective with 18 years of decorated service who suspects the clique is involved in criminal activity.</description>
  <backstory>She lost her partner, DS Morris, during a case three years ago under unexplained circumstances that had supernatural origins she doesn't yet understand.</backstory>
</character>

<character id="4414" name="Eva Kowalski" aliases="Eva, Evie" pronouns="she/her" species="Human" storyRole="Best Friend">
  <eyeColor>green</eyeColor>
  <hair>curly red</hair>
  <height>5'4"</height>
  <age>26</age>
  <glasses>round glasses</glasses>
  <satchel>worn leather satchel full of books</satchel>
  <freckles>freckled complexion</freckles>
  <nervousHabit>tucking hair behind her left ear</nervousHabit>
  <description>Aurora's childhood friend and an occult researcher.</description>
  <backstory>Moved to London two years before Aurora and works as a research assistant at the British Museum's restricted archives. She holds a master's degree in Ancient History from Oxford.</backstory>
</character>

<location id="856" name="The Veil Market" aliases="Veil Market, the Market, underground market" realm="Earth">
  <currentLocation>abandoned Tube station beneath Camden</currentLocation>
  <entryRequirement>bone token</entryRequirement>
  <schedule>moves locations every full moon</schedule>
  <description>A hidden supernatural black market that sells enchanted goods, banned alchemical substances, and information.</description>
</location>

<item id="2004" name="Veil Compass" aliases="the Compass, shadow compass" type="Tool" origin="The Veil Market">
  <casingMaterial>small brass compass</casingMaterial>
  <faceMarkings>face etched with protective sigils</faceMarkings>
  <needle>points toward the nearest supernatural rift or portal</needle>
  <patina>casing has a patina of verdigris</patina>
  <creator>crafted by a Shade artisan</creator>
  <description>A compass attuned to supernatural energy.</description>
</item>
</codex>

Write a fully fleshed out scene of at least 1000 words in the mystery genre.

A detective arrives at a crime scene in an unusual location. Something is wrong — the evidence doesn't add up. A colleague is present, offering a different interpretation. Through careful observation and deduction, the detective begins to see what others have missed.

Use the codex entries above as your world reference for characters and setting.
Only include the prose of the scene without any markdown or commentary.

Model response

The abandoned Tube station swallowed sound like a hungry beast. Detective Harlow Quinn's boots echoed against tiled walls streaked with decades of grime, her flashlight cutting a surgical path through the darkness. Something was off. The crime scene felt... wrong.

Blood spatter traced an impossible pattern across the platform's cracked tiles. Not random. Deliberate. Quinn's sharp jaw clenched as she crouched, her salt-and-pepper hair catching the harsh beam of her light. Her worn leather watch caught a glint of reflection as she traced the blood droplets with a gloved finger.

"Not consistent with a standard assault," she muttered.

Eva Kowalski stood nearby, her round glasses reflecting the forensic lights. Her curly red hair was pulled back, freckled hands clutching a leather satchel that bulged with research materials. "The pattern suggests ritualistic placement," she said, voice tight with academic precision.

Quinn knew Eva wasn't just a research assistant. Her expertise in occult studies ran deeper than the museum's official records. They'd worked together before, though Quinn typically maintained a professional distance.

"Ritualistic how?" Quinn's brown eyes narrowed.

Eva stepped closer, tucking a stray curl behind her left ear—a nervous habit Quinn had noticed before. "The blood droplets form a sigil. Not one I immediately recognize, but definitely not random."

Quinn's military-trained mind processed the information. Eighteen years on the force had taught her to question everything, especially when supernatural markers emerged. Three years ago, her partner DS Morris had disappeared under circumstances that still haunted her nightmares.

She pulled out a small brass compass from her coat pocket. The Veil Compass, its face etched with protective sigils and a patina of verdigris, seemed to vibrate slightly in her hand. The needle twitched, pointing not toward magnetic north, but toward something else entirely.

"This scene isn't just a crime," Quinn said slowly. "It's a message."

Eva nodded, her green eyes intense. "A summoning, perhaps. Or a warning."

The abandoned Tube station felt alive around them, secrets pressed into every crumbling tile and shadowed corner. Something fundamental had been disrupted here—something that went far beyond a simple metropolitan police investigation.

Evaluation details

Result

Evaluator

Details

Meta Data

22.2222%

Adverbs in dialogue tags

Target: ≤10% dialogue tags with adverbs

totalTags

adverbTagCount

adverbTags

0	"Quinn said slowly [slowly]"

43.1818%

AI-ism adverb frequency

Target: <1% AI-ism adverbs (57 tracked)

wordCount

352

totalAiIsmAdverbs

found

adverb	"slightly"
count	1

adverb	"slowly"
count	1

highlights

0	"slightly"
1	"slowly"

100.0000%

AI-ism character names

Target: 0 AI-default names (17 tracked, −20% each)

codexExemptions	(empty)
found	(empty)

100.0000%

AI-ism location names

Target: 0 AI-default location names (33 tracked, −20% each)

codexExemptions	(empty)
found	(empty)

0.0000%

AI-ism word frequency

Target: <1% AI-ism words (140 tracked)

wordCount

352

totalAiIsms

found

word	"glint"
count	1

word	"processed"
count	1

word	"magnetic"
count	1

word	"jaw clenched"
count	1

word	"echoed"
count	1

word	"etched"
count	1

highlights

0	"glint"
1	"processed"
2	"magnetic"
3	"jaw clenched"
4	"echoed"
5	"etched"

66.6667%

Cliche density

Target: ≤1 cliche(s) per 800-word window

totalCliches

maxInWindow

found

label	"eyes widened/narrowed"
count	1

label	"jaw/fists clenched"
count	1

highlights

0	"eyes narrowed"
1	"jaw clenched"

100.0000%

Emotion telling (show vs. tell)

Target: ≤3% sentences with emotion telling

emotionTells	0
narrationSentences	29
matches	(empty)

100.0000%

Filter word density

Target: ≤12% sentences with filter/hedge words

filterCount

hedgeCount

narrationSentences

filterMatches

"watch"

hedgeMatches

0	"seemed to"

100.0000%

Overuse of "that" (subordinate clause padding)

Target: ≤10% sentences with "that" clauses

thatCount	0
totalSentences	32
matches	(empty)

100.0000%

Paragraph length variance

Target: CV ≥0.5 for paragraph word counts

totalParagraphs

mean

28.92

std

14.76

0.51

sampleLengths

0	40
1	50
2	8
3	41
4	31
5	6
6	32
7	38
8	45
9	12
10	12
11	32

87.3563%

Passive voice overuse

Target: ≤5% passive sentences

passiveCount

totalSentences

matches

0	"was pulled"
1	"been disrupted"

100.0000%

Past progressive (was/were + -ing) overuse

Target: ≤10% past progressive verbs

pastProgressiveCount	0
totalVerbs	52
matches	(empty)

100.0000%

Purple prose (modifier overload)

Target: <4% adverbs, <2% -ly adverbs, no adj stacking

wordCount	309
adjectiveStacks	0
stackExamples	(empty)
adverbCount	8
adverbRatio	0.025889967637540454
lyAdverbCount	6
lyAdverbRatio	0.019417475728155338

100.0000%

Repeated phrase echo

Target: ≤20% sentences with echoes (window: 2)

totalSentences	32
echoCount	0
echoWords	(empty)

100.0000%

Sentence length variance

Target: CV ≥0.4 for sentence word counts

totalSentences

mean

10.84

std

5.89

0.543

sampleLengths

0	10
1	22
2	3
3	4
4	1
5	11
6	2
7	1
8	17
9	19
10	8
11	11
12	18
13	12
14	8
15	12
16	11
17	6
18	17
19	15
20	6
21	16
22	16
23	11
24	21
25	13
26	9
27	3
28	6
29	6
30	17
31	15

90.6250%

Sentence opener variety

Target: ≥60% unique sentence openers

consecutiveRepeats	1
diversityRatio	0.5625
totalSentences	32
uniqueOpeners	18

0.0000%

Dialogue tag variety (said vs. fancy)

Target: ≤30% fancy dialogue tags

totalTags

fancyCount

fancyTags

0	"she muttered (mutter)"
1	"she said, voice tight with academic precision (say, voice)"
2	"Quinn said slowly (say slowly)"

77.0619%