Run: 2026-02-19-01KHSZ2GTM4X2VPS4KVEF4VEVJ

Run Details

Model:: Mistral Large 2
Test:: Bad Writing Habits
Scenario:: Mystery: examining a crime scene
Duration:: 16.3s
Cost:: $0.006984

Evaluation summary

Score	Evaluator	Details Link
100.00%	Adverbs in dialogue tags	(details)
82.91%	AI-ism adverb frequency	(details)
100.00%	AI-ism character names	(details)
100.00%	AI-ism location names	(details)
65.81%	AI-ism word frequency	(details)
100.00%	Cliché density	(details)
100.00%	Emotion telling (show vs. tell)	(details)
100.00%	Filter word density	(details)
100.00%	Gibberish response detection	(details)
100.00%	Markdown formatting overuse	(details)
100.00%	Missing dialogue indicators (quotation marks)	(details)
32.08%	Name drop frequency	(details)
0.00%	Narrator intent-glossing	(details)
100.00%	"Not X but Y" pattern overuse	(details)
100.00%	Overuse of "that" (subordinate clause padding)	(details)
100.00%	Paragraph length variance	(details)
91.50%	Passive voice overuse	(details)
100.00%	Past progressive (was/were + -ing) overuse	(details)
56.28%	Em-dash & semicolon overuse	(details)
100.00%	Purple prose (modifier overload)	(details)
100.00%	Repeated phrase echo	(details)
100.00%	Sentence length variance	(details)
64.14%	Sentence opener variety	(details)
100.00%	Adverb-first sentence starts	(details)
100.00%	Pronoun-first sentence starts	(details)
34.47%	Subject-first sentence starts	(details)
0.00%	Subordinate conjunction sentence starts	(details)
95.24%	Technical jargon density	(details)
100.00%	Useless dialogue additions	(details)
100.00%	Dialogue tag variety (said vs. fancy)	(details)
84.0807%	Overall

Model input

System

You are an expert fiction writer.

Always keep the following rules in mind:
- Write in past tense and use General English spelling, grammar, and colloquialisms/slang.
- Write in active voice
- Always follow the "show, don't tell" principle.
- Avoid adverbs and cliches and overused/commonly used phrases. Aim for fresh and original descriptions.
- Convey events and story through dialogue.
- Mix short, punchy sentences with long, descriptive ones. Drop fill words to add variety.
- Skip "he/she said said" dialogue tags and convey people's actions or face expressions through their speech
- Avoid mushy dialog and descriptions, have dialogue always continue the action, never stall or add unnecessary fluff. Vary the descriptions to not repeat yourself.
- Put dialogue on its own paragraph to separate scene and action.
- Reduce indicators of uncertainty like "trying" or "maybe"

When writing text:
- NEVER conclude the scene on your own, follow the beat instructions very closely.
- NEVER end with foreshadowing.
- NEVER write further than what I prompt you with.
- AVOID imagining possible endings, NEVER deviate from the instructions.
- STOP EARLY if the continuation contains what was required in the instructions. You do not need to fill out the full amount of words possible.

User/Human

Take into account the following glossary of characters/locations/items/lore... when writing your response:
<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>

Continue the story using the following instructions:
<instructions>
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.
</instructions>

Model response

The abandoned Tube station smelled of damp concrete and rusted iron. Quinn stepped over the police tape, her boots clicking against the cracked tiles. The beam of her torch cut through the gloom, illuminating the graffiti-scrawled walls—tags that looked too deliberate, too precise, like symbols rather than vandalism.

"Quinn. Over here."

Detective Sergeant Cole stood near the edge of the platform, his hands shoved deep in his pockets. He nodded toward the body sprawled across the tracks. A man in a cheap suit, his tie still knotted tight, one arm flung out as if reaching for something just beyond his fingertips.

She crouched, her knees popping. The victim’s fingers were clean, no dirt under the nails. No defensive wounds. "Time of death?"

"ME’s initial estimate puts it around midnight. No signs of struggle, no forced entry." Cole shrugged. "Could be a jumper. Or a push."

Quinn’s torch flickered over the man’s face. His eyes were open, fixed in an expression that wasn’t quite fear, wasn’t quite surprise. More like recognition. She reached out, tilting his head with a gloved finger. The skin at his temple was unmarked, but the hair there was matted, sticky.

"Blood," she said.

Cole exhaled through his nose. "Scalp wound. Could’ve hit his head on the way down."

"Not unless he fell backward." She stood, sweeping the beam across the platform. The tiles were dry. No blood spatter. No drag marks. "He didn’t jump. And he didn’t fall."

Cole’s jaw tightened. "You’re saying someone carried him down here, laid him out like this, and then what? Left no trace?"

Quinn didn’t answer. Her torch caught on something half-buried in the gravel between the tracks. She stepped down, careful not to disturb the scene, and plucked it free. A small brass compass, its casing tarnished green. The needle spun lazily, refusing to settle.

Cole frowned. "What’s that?"

"Nothing." She pocketed it. "Bag the body. I want the ME to check for traces of anything unusual—drugs, toxins, anything that might explain why he’s not bruised."

Cole’s radio crackled. He turned away, muttering into it. Quinn took the moment to examine the compass again. The sigils etched into its face were worn but unmistakable—wards, protection marks. Not something you’d find in a tourist shop.

A shadow moved at the edge of her vision. She turned, torch raised, but there was nothing. Just the empty tunnel, the distant drip of water.

Cole turned back. "ME’s on his way. You want to tell me what’s really going on?"

Quinn weighed the compass in her palm. "You ever hear of the Veil Market?"

Cole blinked. "The what now?"

"Underground black market. Moves every full moon. Sells things that aren’t exactly legal." She snapped the compass shut. "And things that aren’t exactly *real*."

Cole’s laugh was sharp. "You’re kidding, right? Next you’re gonna tell me this guy was into witchcraft."

Quinn’s gaze flicked back to the body. The man’s shoes were polished, his cufflinks real silver. Not the kind of person who’d end up in a place like this by accident. "I’m not kidding."

A gust of wind howled through the tunnel, carrying the scent of ozone and something older, something like burnt copper. The compass in her pocket grew warm.

Cole’s radio hissed static. "Quinn, you feel that?"

She didn’t answer. The needle on the compass had stopped spinning. It pointed straight at the far wall, where the graffiti coiled into shapes that now looked less like tags and more like warnings.

Evaluation details

Result

Evaluator

Details

Meta Data

100.00%

Adverbs in dialogue tags

Target: ≤10% dialogue tags with adverbs

totalTags	5
adverbTagCount	0
adverbTags	(empty)
dialogueSentences	20
tagDensity	0.25
leniency	0.5
rawRatio	0
effectiveRatio	0

82.91%

AI-ism adverb frequency

Target: <2% AI-ism adverbs (58 tracked)

wordCount

585

totalAiIsmAdverbs

found

adverb	"lazily"
count	1

adverb	"really"
count	1

highlights

0	"lazily"
1	"really"

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)

65.81%

AI-ism word frequency

Target: <2% AI-ism words (290 tracked)

wordCount

585

totalAiIsms

found

word	"gloom"
count	1

word	"flickered"
count	1

word	"etched"
count	1

word	"flicked"
count	1

highlights

0	"gloom"
1	"flickered"
2	"etched"
3	"flicked"

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	51
matches	(empty)

100.00%

Filter word density

Target: ≤3% sentences with filter/hedge words

filterCount	0
hedgeCount	0
narrationSentences	51
filterMatches	(empty)
hedgeMatches	(empty)

100.00%

Gibberish response detection

Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)

analyzedSentences	66
gibberishSentences	0
adjustedGibberishSentences	0
longSentenceCount	0
runOnParagraphCount	0
giantParagraphCount	0
wordSaladCount	0
repetitionLoopCount	0
controlTokenCount	0
maxSentenceWordsSeen	26
ratio	0
matches	(empty)

100.00%

Markdown formatting overuse

Target: ≤5% words in markdown formatting

markdownSpans

markdownWords

totalWords

580

ratio

0.002

matches

"real"

100.00%

Missing dialogue indicators (quotation marks)

Target: ≤10% speech attributions without quotation marks

totalAttributions	9
unquotedAttributions	0
matches	(empty)

32.08%

Name drop frequency

Target: ≤1.0 per-name mentions per 100 words

totalMentions

wordCount

424

uniqueNames

maxNameDensity

2.36

worstName

"Cole"

maxWindowNameDensity

3.5

worstWindowName

"Cole"

discoveredNames

Tube	1
Sergeant	1
Cole	10
Quinn	6

persons

0	"Sergeant"
1	"Cole"
2	"Quinn"

places

(empty)

globalScore

0.321

windowScore

0.5

0.00%

Narrator intent-glossing

Target: ≤2% narration sentences with intent-glossing patterns

analyzedSentences

glossingSentenceCount

matches

0	"as if reaching for something just beyond his fingertips"
1	"quite fear, wasn’t quite surprise"
2	"quite surprise"
3	"something like burnt copper"

100.00%

"Not X but Y" pattern overuse

Target: ≤1 "not X but Y" per 1000 words

totalMatches	0
per1kWords	0
wordCount	580
matches	(empty)

100.00%

Overuse of "that" (subordinate clause padding)

Target: ≤2% sentences with "that" clauses

thatCount	0
totalSentences	66
matches	(empty)

100.00%

Paragraph length variance

Target: CV ≥0.5 for paragraph word counts

totalParagraphs

mean

24.17

std

14.29

0.591

sampleLengths

0	48
1	3
2	50
3	21
4	23
5	49
6	3
7	15
8	30
9	21
10	43
11	4
12	27
13	38
14	26
15	16
16	14
17	5
18	24
19	17
20	34
21	27
22	8
23	34

91.50%

Passive voice overuse

Target: ≤2% passive sentences

passiveCount

totalSentences

matches

0	"was matted"
1	"were polished"

100.00%

Past progressive (was/were + -ing) overuse

Target: ≤2% past progressive verbs

pastProgressiveCount	0
totalVerbs	82
matches	(empty)

56.28%

Em-dash & semicolon overuse

Target: ≤2% sentences with em-dashes/semicolons

emDashCount

semicolonCount

flaggedSentences

totalSentences

ratio

0.03

matches

0	"The beam of her torch cut through the gloom, illuminating the graffiti-scrawled walls—tags that looked too deliberate, too precise, like symbols rather than vandalism."
1	"The sigils etched into its face were worn but unmistakable—wards, protection marks."

100.00%

Purple prose (modifier overload)

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

wordCount	428
adjectiveStacks	0
stackExamples	(empty)
adverbCount	15
adverbRatio	0.035046728971962614
lyAdverbCount	1
lyAdverbRatio	0.002336448598130841

100.00%

Repeated phrase echo

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

totalSentences	66
echoCount	0
echoWords	(empty)

100.00%

Sentence length variance

Target: CV ≥0.4 for sentence word counts

totalSentences

mean

8.79

std

5.9

0.672

sampleLengths

0	11
1	13
2	24
3	3
4	17
5	9
6	24
7	5
8	10
9	3
10	3
11	16
12	7
13	7
14	15
15	3
16	10
17	14
18	3
19	5
20	10
21	13
22	4
23	3
24	3
25	7
26	3
27	18
28	3
29	12
30	13
31	8
32	7
33	2
34	2
35	4
36	23
37	3
38	6
39	9
40	12
41	8
42	9
43	8
44	9
45	3
46	13
47	7
48	7
49	2

64.14%

Sentence opener variety

Target: ≥60% unique sentence openers

consecutiveRepeats	1
diversityRatio	0.3939393939393939
totalSentences	66
uniqueOpeners	26

100.00%

Adverb-first sentence starts

Target: ≥3% sentences starting with an adverb

adverbCount

totalSentences

matches

0	"More like recognition."
1	"Just the empty tunnel, the"

ratio

0.043

100.00%

Pronoun-first sentence starts

Target: ≤30% sentences starting with a pronoun

pronounCount

totalSentences

matches

0	"He nodded toward the body"
1	"She crouched, her knees popping."
2	"His eyes were open, fixed"
3	"She reached out, tilting his"
4	"She stood, sweeping the beam"
5	"Her torch caught on something"
6	"She stepped down, careful not"
7	"She pocketed it"
8	"He turned away, muttering into"
9	"She turned, torch raised, but"
10	"She snapped the compass shut"
11	"She didn’t answer."
12	"It pointed straight at the"

ratio

0.277

34.47%

Subject-first sentence starts

Target: ≤72% sentences starting with a subject

subjectCount

totalSentences

matches

0	"The abandoned Tube station smelled"
1	"Quinn stepped over the police"
2	"The beam of her torch"
3	"Detective Sergeant Cole stood near"
4	"He nodded toward the body"
5	"A man in a cheap"
6	"She crouched, her knees popping."
7	"The victim’s fingers were clean,"
8	"Quinn’s torch flickered over the"
9	"His eyes were open, fixed"
10	"She reached out, tilting his"
11	"The skin at his temple"
12	"Cole exhaled through his nose."
13	"She stood, sweeping the beam"
14	"The tiles were dry."
15	"Cole’s jaw tightened."
16	"Quinn didn’t answer."
17	"Her torch caught on something"
18	"She stepped down, careful not"
19	"A small brass compass, its"

ratio

0.851

0.00%

Subordinate conjunction sentence starts

Target: ≥2% sentences starting with a subordinating conjunction

subConjCount	0
totalSentences	47
matches	(empty)
ratio	0

95.24%

Technical jargon density

Target: ≤6% sentences with technical-jargon patterns

analyzedSentences

technicalSentenceCount

matches

0	"A man in a cheap suit, his tie still knotted tight, one arm flung out as if reaching for something just beyond his fingertips."

100.00%

Useless dialogue additions

Target: ≤5% dialogue tags with trailing filler fragments

totalTags	5
uselessAdditionCount	0
matches	(empty)

100.00%

Dialogue tag variety (said vs. fancy)

Target: ≤10% fancy dialogue tags

totalTags

fancyCount

fancyTags

0	"She snapped (snap)"

dialogueSentences

tagDensity

0.1

leniency

0.2

rawRatio

0.5

effectiveRatio

0.1

84.0807%