NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M36NMB947FWVG17QA4A0RR70
Run Details
Model:
GPT-6 Luna
Test:
Bad Writing Habits
Scenario:
Mystery: examining a crime scene
Duration:
36.9s
Cost:
$0.001460
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Adverbs in dialogue tags
(details)
81.25%
AI-ism adverb frequency
(details)
100.00%
AI-ism character names
(details)
100.00%
AI-ism location names
(details)
81.25%
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)
66.67%
Name drop frequency
(details)
100.00%
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)
90.29%
Passive voice overuse
(details)
100.00%
Past progressive (was/were + -ing) overuse
(details)
100.00%
Em-dash & semicolon overuse
(details)
100.00%
Purple prose (modifier overload)
(details)
100.00%
Repeated phrase echo
(details)
100.00%
Sentence length variance
(details)
50.75%
Sentence opener variety
(details)
87.72%
Adverb-first sentence starts
(details)
100.00%
Pronoun-first sentence starts
(details)
38.95%
Subject-first sentence starts
(details)
32.89%
Subordinate conjunction sentence starts
(details)
100.00%
Technical jargon density
(details)
100.00%
Useless dialogue additions
(details)
100.00%
Dialogue tag variety (said vs. fancy)
(details)
90.9919%
Overall
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
Messages
Meta
AI
By the time Detective Harlow
Quinn reached the bottom of the station stairs, the dead man’s
watch
had stopped at 11:17 and every clock in the abandoned platform showed a different hour.
A constable held the cordon
at the ticket gate, where a strip of blue-and-white tape sagged between two rusted pillars. Beyond it, the old Camden station breathed damp air.
The tunnel lamps had been
dead for decades, but someone had strung portable work lights along the platform.
Their hard glare flattened the
scene: peeling tiles, blackened rails, a body laid out on the yellow safety line. Quinn ducked beneath the tape.
She wore her coat buttoned
to the throat, her movements spare and exact.
The worn leather watch on
her left wrist showed 2:08.
She checked it twice before
looking at the others. “Who found him?” “Market security,” said Sergeant Vale. “One of their people. He says the body was there when they opened the platform.” “Opened it for what?”
Vale gave a short, humorless
shrug. “Apparently there’s a market.” Quinn looked past him.
The platform widened at its
far end into a shadowed concourse cluttered with folding stalls, crates, and canvas awnings.
Someone had packed up in
haste.
A string of paper lanterns
swung
gently
though the air was still. There were no signs, no advertisements, no ordinary rubbish.
The place looked less abandoned
than temporarily forgotten. “An underground market,” she said. “Unofficial,” Vale replied. “The security man called it the Veil Market. Said it moves every full moon. He’s been taken upstairs for a statement.” Quinn studied him. “And you believed him?” “I believe there’s a dead man and a lot of people who don’t want to talk. I’m keeping an open mind.” It was a good answer.
Vale had learned not to
mistake disbelief for caution.
The victim was in his
thirties, lean, dark-haired, dressed in a charcoal suit too fine for the station. No visible wound.
His face had the waxy
stillness of sudden death. One hand rested beside his hip; the other
was clenched
against his chest, fingers folded around nothing. Quinn crouched without touching him.
The scene smelled of wet
stone, cold metal, and something
sweetly
medicinal. A faint white dust clung to the man’s cuffs. “Any identification?” “Wallet’s on the table.” Vale indicated a folding stall nearby. “Edwin Marsh. Accountant. No known connection to this place.” “Cause of death?” “Medic says cardiac arrest. No marks. No blood. No sign of a struggle.” Quinn glanced at the
watch
on the dead man’s wrist. The glass
was cracked
, the hands fixed at 11:17. “When was he found?” “Just before one.” “Then why did the
watch
stop at 11:17?” “Impact, maybe. He could’ve fallen.” “He’s lying on a flat platform, and his clothes aren’t scuffed.” Vale stood over her, arms crossed. “You think someone arranged him?” “I think the scene is telling us something. We haven’t worked out what.”
She rose and let her
gaze travel over the platform. The nearest stall had
been stripped
bare, but a dozen small objects remained on the counter: a copper spoon, a lump of black glass, a coil of wire, and a brass compass green with age.
Its face was etched with
marks she didn’t recognize. A needle
trembled
beneath the glass. Quinn leaned closer. The needle swung once, then settled—
not north, but toward the tunnel entrance
. “What’s that?” Vale asked. “An item somebody forgot.” “Or planted.” “Could be.”
She took a photograph with
her phone, then examined the stall’s surface. Dust lay in a soft gray film everywhere except for a clean oval where something had sat. A narrow trail cut through the dust from that oval to the platform edge. It ended beside Marsh’s shoes. “Do we know if the stalls were here when the body was found?” “The security man says the market was open. He claims the traders vanished before the police arrived.” “Vanished how?” “He says they left.” “Convenient.” Vale nodded toward the victim. “There’s another problem. The scene technician found a partial shoeprint near the body. It points toward the track.” Quinn followed his gaze. The print lay in a wet patch beside the yellow line: the
sharply
defined tread of a right shoe, heel toward the platform, toe toward the rails.
It was the sort of
impression a fleeing killer might leave. “Any matching prints on the stairs?” she asked. “None.” “On the platform?” “Only the victim’s shoes and the security man’s. We’ve checked the obvious route.” “Obvious to whom?” Vale frowned. Quinn stepped around the print. The platform tiles were damp, but the water had gathered in shallow depressions rather than spread evenly. She looked up. The ceiling was dry. No leak overhead. A slick shine coated the tile near the track, ending abruptly at the yellow line. “Was this wet when you got here?” “According to the first officer, yes.” “Where’s the rest of it?” Vale followed her gaze. “What rest?” “Water. If someone walked through a puddle, there should be more than one print. If he stepped out of the train tunnel, there should be mud. If the body was dragged, there should be a smear. We have a single wet impression surrounded by dry dust.” “Could’ve
been made
with a wet shoe.” “Then whoever made it stood in exactly one spot, took one step toward the track, and evaporated.” Vale’s mouth tightened. “You’re suggesting the print was faked.” “I’m suggesting it doesn’t belong to the same world as the rest of this scene.” The words sounded melodramatic. She disliked them.
She disliked more that they
felt accurate. She studied the print again. The tread was deep at the toe and shallow at the heel, as if
weight
had landed hard on the front of the foot. The water inside it was still clear. No grit, no oil, no residue from the rails. A step toward the track. A deliberate arrow. “Maybe the murderer wanted us looking down the tunnel,” Vale said. “That’s one interpretation.” “And yours?” Quinn looked at the body. “That it wasn’t made by a person walking.”
She heard the silence that
followed.
Somewhere in the concourse, one
of the paper lanterns tapped
lightly
against its string. Vale lowered his voice. “You’ve seen something like this before?” “No.” Three years ago, her partner had disappeared from a locked room with a dead suspect, two spent cartridges, and a smear of ash across the floor. There had been no windows. The ash had led nowhere. The official report had blamed a trapdoor no one could find and an error no one could explain. Quinn had never told Vale that sometimes she still woke with Morris’s name in her mouth.
She looked away from the
track. “Bag the compass,” she said. “Carefully. Don’t turn the needle.” “You think it’s relevant?” “I think it’s the only thing here that appears to be pointing at something.” A gloved technician approached with an evidence bag. Quinn held up a hand. “Before you lift it, photograph the face from above. And the casing. Get the markings.” The compass needle shivered against the glass. For a moment, the
etched
sigils
seemed to
darken.
Then the point swung sharply
away from the tunnel and aimed at the body. Quinn felt the muscles in her jaw lock. Vale had seen it too. “It moved.” “Yes.” “Could be magnetized.” “Maybe.”
She made herself inspect the
scene rather than the instrument. The compass could be a trick. There were magnets, wires, every kind of cheap mechanism.
She needed facts that didn’t
depend on believing the impossible. “Where’s the medic’s report?” “Coming.” “Has anyone checked for needle marks?” “Not yet.” “Check the neck, the inside of the elbows, the gums. Toxicology as soon as we can get the body out.” Vale looked from the compass to the white dust on Marsh’s cuffs. “You think he was poisoned?” “I think his heart stopped without leaving a mark, and there’s powder on his clothes. That’s enough to ask.” She moved to the stall. The oval in the dust was nearly circular, perhaps the size of a bowl. A thin crescent of dark residue remained at one edge. Quinn rubbed the air above it, resisting the urge to touch. The sweet medicinal scent was stronger here. “Any containers found?” “Nothing with the body.” “Then somebody took one.” “Or it was never there.” Quinn turned to face the platform again. The evidence had arranged itself into a story: Marsh came to the market. He stood at the stall. Something happened at 11:17. He collapsed on the line. A shoeprint suggested a struggle or escape into the tunnel. But the body was clean, the floor mostly dry, the lone print too
perfect
. And the compass, whether
mechanical
or not, pointed to him. The story didn’t add up because someone had written it backward. She walked to the body and crouched beside his left hand. His fingers
were curled
tightly
inward. She could see a faint crescent of dirt beneath the nails, and a small bruise on the heel of his palm. Not from a fall, she thought. Too localized. The bruise formed a pale ring, as if something narrow had pressed there. “What did he have in his hand?” she asked. “Nothing,” Vale said. “We checked.” Quinn examined the fingers. The tips were clean, but a faint black line crossed the skin beneath the index finger. She took a close photograph. “Could be ink.” “From the stall?” “Maybe.” The dead man’s right hand lay open, palm upward. In the center of it sat a tiny cut, too neat for an accident. Quinn leaned in. A bead of dried blood had run toward the wrist, but stopped at the crease. “Did he bleed anywhere else?” “No.” “Then the cut happened while he was standing.” Vale stared at her. “How can you tell?” “Because the blood didn’t run down his fingers.” She looked at the sole of Marsh’s right shoe. A faint smear of the same white dust streaked its edge. On the left shoe, none. The wet print beside the track had the same angular tread, but it pointed the wrong way. It had
been made
by a right shoe. Marsh wore polished leather oxfords with smooth soles. Quinn stood
slowly
. “The print isn’t a shoeprint,” she said. “Not a human shoe. It’s a cast, or a stamp. Whoever made it wanted us to imagine a person running toward the tunnel.” Vale glanced toward the stairs. “So the killer left no footprints.” “Or didn’t need to walk here.” The compass needle remained fixed on the body. The market security man had said the stalls vanished. Perhaps they had packed away.
Perhaps they had never been
there until someone needed them. Quinn looked at the stripped counter, the dust-free oval, the single cut in Marsh’s palm. A transaction, not an ambush. Something had changed hands. Something had
been opened
, spilled, or broken. She bent over the victim once more. The stain beneath his cuffs caught the work light. It wasn’t dust. The particles glittered faintly, like powdered glass. “Vale, get everyone off this platform. No one touches the stalls. Have the techs collect air samples, residue, and the water around that print. And find the security man’s exact words about how Marsh was discovered.” “You think he’s lying?” “I think he saw something and gave us the version he thought we could accept.” “And what version is that?” Quinn looked at the compass. Its needle had begun to tremble again, though nothing nearby had moved. It pointed past Marsh now, toward the blank tiled wall behind the stalls. She had spent eighteen years
learning that every
scene had an answer. Evidence might be concealed, contaminated, or misread, but it obeyed rules. A person had to cross a room to enter it. A body had to bear the consequences of whatever killed it. Time moved forward. The
watch
at 11:17 had stopped beneath her gaze. Her own
watch
continued to tick. “I don’t know yet,” she said. “But I know it’s not the story we’ve been given.” She turned toward the wall. The tiles there looked no different from the rest: cream glaze, hairline cracks, old soot. Yet the compass needle held steady, aimed at a point just above the platform, where a hair-thin seam ran between two tiles. Quinn stepped close. The seam was warm. Behind her, Vale drew a sharp breath. Quinn didn’t touch the wall. She took one step back and raised her hand to stop the others from coming nearer. In the narrow seam, something dark
pulsed
once, like a shadow crossing the far side of a door. Her worn
watch
ticked against her wrist. From the tunnel came the sound of a train arriving, though the rails lay empty and cold.
created_at
"2026-09-23T08:20:34.218Z"
9/23/2026, 8:20:34 AM
model
"openai/gpt-6-luna-20260922"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1352
moderation_latency
(null)
generation_time
36831
tokens_prompt
896
tokens_completion
3340
native_tokens_prompt
835
native_tokens_completion
2754
native_tokens_completion_images
0
native_tokens_reasoning
0
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"completed"
service_tier
"default"
usage
0.0014605
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790151634-HsdN1ms3EC5im6d7kIo5"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790151634-mNcIqLxa5Cx4o6cCpVs9"
upstream_id
"resp_0883cf6334a2dc71016ab38bd24df887d19dcd1eb1a0bf1381"
provider_responses
0
endpoint_id
"05e94e02-b9c2-4bbb-ba55-4082ee9ad687"
id
"resp_0883cf6334a2dc71016ab38bd24df887d19dcd1eb1a0bf1381"
is_byok
false
latency
303
model_permaslug
"openai/gpt-6-luna-20260922"
provider_name
"OpenAI"
status
200
total_cost
0.0014605
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Adverbs in dialogue tags
Target: ≤10% dialogue tags with adverbs
totalTags
11
adverbTagCount
0
adverbTags
(empty)
dialogueSentences
93
tagDensity
0.118
leniency
0.237
rawRatio
0
effectiveRatio
0
81.25%
AI-ism adverb frequency
Target: <2% AI-ism adverbs (58 tracked)
wordCount
2133
totalAiIsmAdverbs
8
found
0
adverb
"gently"
count
1
1
adverb
"sweetly"
count
1
2
adverb
"sharply"
count
2
3
adverb
"lightly"
count
1
4
adverb
"carefully"
count
1
5
adverb
"tightly"
count
1
6
adverb
"slowly"
count
1
highlights
0
"gently"
1
"sweetly"
2
"sharply"
3
"lightly"
4
"carefully"
5
"tightly"
6
"slowly"
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)
81.25%
AI-ism word frequency
Target: <2% AI-ism words (290 tracked)
wordCount
2133
totalAiIsms
8
found
0
word
"etched"
count
2
1
word
"trembled"
count
1
2
word
"weight"
count
1
3
word
"silence"
count
1
4
word
"perfect"
count
1
5
word
"mechanical"
count
1
6
word
"pulsed"
count
1
highlights
0
"etched"
1
"trembled"
2
"weight"
3
"silence"
4
"perfect"
5
"mechanical"
6
"pulsed"
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
164
matches
(empty)
100.00%
Filter word density
Target: ≤3% sentences with filter/hedge words
filterCount
2
hedgeCount
1
narrationSentences
164
filterMatches
0
"watch"
hedgeMatches
0
"seemed to"
100.00%
Gibberish response detection
Target: ≤1% gibberish-like sentences (hard fail if a sentence exceeds 800 words)
analyzedSentences
245
gibberishSentences
0
adjustedGibberishSentences
0
longSentenceCount
0
runOnParagraphCount
0
giantParagraphCount
0
wordSaladCount
0
repetitionLoopCount
0
controlTokenCount
0
repeatedSegmentCount
0
maxSentenceWordsSeen
46
ratio
0
matches
(empty)
100.00%
Markdown formatting overuse
Target: ≤5% words in markdown formatting
markdownSpans
0
markdownWords
0
totalWords
2132
ratio
0
matches
(empty)
100.00%
Missing dialogue indicators (quotation marks)
Target: ≤10% speech attributions without quotation marks
totalAttributions
23
unquotedAttributions
0
matches
(empty)
66.67%
Name drop frequency
Target: ≤1.0 per-name mentions per 100 words
totalMentions
54
wordCount
1465
uniqueNames
8
maxNameDensity
1.5
worstName
"Quinn"
maxWindowNameDensity
3
worstWindowName
"Vale"
discoveredNames
Detective
1
Harlow
1
Quinn
22
Camden
1
Sergeant
1
Vale
20
Marsh
7
Morris
1
persons
0
"Harlow"
1
"Quinn"
2
"Sergeant"
3
"Vale"
4
"Marsh"
5
"Morris"
places
(empty)
globalScore
0.749
windowScore
0.667
100.00%
Narrator intent-glossing
Target: ≤2% narration sentences with intent-glossing patterns
analyzedSentences
103
glossingSentenceCount
0
matches
(empty)
100.00%
"Not X but Y" pattern overuse
Target: ≤1 "not X but Y" per 1000 words
totalMatches
1
per1kWords
0.469
wordCount
2132
matches
0
"not north, but toward the tunnel entrance"
100.00%
Overuse of "that" (subordinate clause padding)
Target: ≤2% sentences with "that" clauses
thatCount
1
totalSentences
245
matches
0
"learning that every"
100.00%
Paragraph length variance
Target: CV ≥0.5 for paragraph word counts
totalParagraphs
130
mean
16.4
std
17.07
cv
1.041
sampleLengths
0
32
1
66
2
37
3
3
4
20
5
4
6
10
7
58
8
5
9
24
10
7
11
21
12
14
13
47
14
26
15
2
16
19
17
3
18
13
19
19
20
4
21
3
22
8
23
5
24
11
25
11
26
13
27
60
28
15
29
4
30
4
31
2
32
2
33
48
34
13
35
17
36
2
37
4
38
1
39
23
40
42
41
8
42
1
43
3
44
13
45
3
46
2
47
47
48
7
49
6
90.29%
Passive voice overuse
Target: ≤2% passive sentences
passiveCount
7
totalSentences
164
matches
0
"was clenched"
1
"was cracked"
2
"been stripped"
3
"was etched"
4
"were curled"
5
"been made"
6
"been opened"
100.00%
Past progressive (was/were + -ing) overuse
Target: ≤2% past progressive verbs
pastProgressiveCount
0
totalVerbs
245
matches
(empty)
100.00%
Em-dash & semicolon overuse
Target: ≤2% sentences with em-dashes/semicolons
emDashCount
1
semicolonCount
1
flaggedSentences
2
totalSentences
245
ratio
0.008
matches
0
"One hand rested beside his hip; the other was clenched against his chest, fingers folded around nothing."
1
"The needle swung once, then settled—not north, but toward the tunnel entrance."
100.00%
Purple prose (modifier overload)
Target: <4% adverbs, <2% -ly adverbs, no adj stacking
wordCount
1471
adjectiveStacks
0
stackExamples
(empty)
adverbCount
53
adverbRatio
0.03602991162474507
lyAdverbCount
13
lyAdverbRatio
0.008837525492861998
100.00%
Repeated phrase echo
Target: ≤20% sentences with echoes (window: 2)
totalSentences
245
echoCount
0
echoWords
(empty)
100.00%
Sentence length variance
Target: CV ≥0.4 for sentence word counts
totalSentences
245
mean
8.7
std
6.54
cv
0.752
sampleLengths
0
32
1
20
2
9
3
18
4
19
5
5
6
13
7
10
8
9
9
3
10
5
11
15
12
4
13
6
14
4
15
4
16
19
17
6
18
12
19
9
20
8
21
5
22
3
23
21
24
3
25
4
26
21
27
5
28
9
29
18
30
3
31
9
32
17
33
5
34
12
35
9
36
2
37
10
38
9
39
3
40
13
41
10
42
9
43
4
44
3
45
8
46
5
47
11
48
6
49
5
50.75%
Sentence opener variety
Target: ≥60% unique sentence openers
consecutiveRepeats
9
diversityRatio
0.32653061224489793
totalSentences
245
uniqueOpeners
80
87.72%
Adverb-first sentence starts
Target: ≥3% sentences starting with an adverb
adverbCount
4
totalSentences
152
matches
0
"Somewhere in the concourse, one"
1
"Then the point swung sharply"
2
"Perhaps they had packed away."
3
"Perhaps they had never been"
ratio
0.026
100.00%
Pronoun-first sentence starts
Target: ≤30% sentences starting with a pronoun
pronounCount
36
totalSentences
152
matches
0
"Their hard glare flattened the"
1
"She wore her coat buttoned"
2
"She checked it twice before"
3
"It was a good answer."
4
"His face had the waxy"
5
"She rose and let her"
6
"Its face was etched with"
7
"She took a photograph with"
8
"It ended beside Marsh’s shoes."
9
"It was the sort of"
10
"She looked up."
11
"She disliked them."
12
"She disliked more that they"
13
"She studied the print again."
14
"She heard the silence that"
15
"She looked away from the"
16
"She made herself inspect the"
17
"She needed facts that didn’t"
18
"She moved to the stall."
19
"He stood at the stall."
ratio
0.237
38.95%
Subject-first sentence starts
Target: ≤72% sentences starting with a subject
subjectCount
128
totalSentences
152
matches
0
"A constable held the cordon"
1
"The tunnel lamps had been"
2
"Their hard glare flattened the"
3
"Quinn ducked beneath the tape."
4
"She wore her coat buttoned"
5
"The worn leather watch on"
6
"She checked it twice before"
7
"Vale gave a short, humorless"
8
"Quinn looked past him."
9
"The platform widened at its"
10
"Someone had packed up in"
11
"A string of paper lanterns"
12
"The place looked less abandoned"
13
"Quinn studied him."
14
"It was a good answer."
15
"Vale had learned not to"
16
"The victim was in his"
17
"His face had the waxy"
18
"Quinn crouched without touching him."
19
"The scene smelled of wet"
ratio
0.842
32.89%
Subordinate conjunction sentence starts
Target: ≥2% sentences starting with a subordinating conjunction
subConjCount
1
totalSentences
152
matches
0
"By the time Detective Harlow"
ratio
0.007
100.00%
Technical jargon density
Target: ≤6% sentences with technical-jargon patterns
analyzedSentences
60
technicalSentenceCount
0
matches
(empty)
100.00%
Useless dialogue additions
Target: ≤5% dialogue tags with trailing filler fragments
totalTags
11
uselessAdditionCount
0
matches
(empty)
100.00%
Dialogue tag variety (said vs. fancy)
Target: ≤10% fancy dialogue tags
totalTags
10
fancyCount
0
fancyTags
(empty)
dialogueSentences
93
tagDensity
0.108
leniency
0.215
rawRatio
0
effectiveRatio
0
90.9919%