Quick look above; the showcases render the full Slipstream bank with sample data. Captured values only on A-side — anatomy stays ours until we craft a library.
Style Guide preset (skin)
primary
#4cba74
pri-accent
#25294d
secondary
—
sec-accent
—
Surfaces — page · alt · emphasis · inverse
page
#ffffff
alt
—
emphasis
—
inverse
—
btn hover
#3eaa62
Heading font
Instrument Sans (Google)
Body font
Instrument Sans (Google)
Heading line-height
0.9
Mobile h1
52px → 34px
Mobile overflow
none
Base size
18px
Type scale
not detected
Rounding
Very rounded (50px)
Line width
not detected
Button hover
brighten background
Text-over-image halo
not detected
Heading tracking
not detected
Label tracking
not detected
Focus ring
not detected
Easing
not detected
Container max
1240px
Section padding
not detected
Accent gradient
not detected
Icon stroke
not detected
Spacing unit
not detected
Transition
300ms
Link underline
false
Button
not detected
rendered (getComputedStyle on laid-out page) — higher fidelity than the regex engine
Capture audit — what's real vs default
12 of 27 values captured from this site. The rest fall back to library defaults — shown so nothing hardcoded can pass as captured.
Brand primarycaptured: #4cba74
Secondary colourdefault
Page backgroundcaptured: #ffffff
Text colourdefault
Text colour (inverse)default
Heading colourdefault
Link colourdefault
Link colour (inverse)default
Heading fontcaptured: Instrument Sans
Body fontcaptured: Instrument Sans
Heading size (H1)captured: 59px
Heading weightcaptured: 500
Heading trackingdefault
Primary button bgcaptured: #4cba74
Primary button radiuscaptured: 50
Roundingcaptured: 50px
Line / border widthdefault
Button hoverdefault
Focus ringdefault
Easingdefault
Transition speedcaptured: 300ms
Container max-widthcaptured: 1240px
Section paddingdefault
Accent gradientdefault
Image treatmentcaptured: captured
Text-over-image halodefault
Icon strokedefault
Alt background (derived from page)library default
Card paddinglibrary default
Small/large shadow scalelibrary default
Secondary-button outline stylelibrary default
Inventory (raw evidence)
Palette by frequency
#ffffff ·866
#000000 ·550
#85ba35 ·115
#25294d ·94
#333333 ·68
#396ab1 ·62
#85ba34 ·61
#fdef70 ·60
#76a1de ·56
#eeeeee ·51
#f7f7fa ·47
#aaaaaa ·43
#7eb939 ·42
#efce4a ·41
#33373d ·35
#cccccc ·34
#999999 ·28
#cc3366 ·28
Radii
0px2px3px4px5px6px8px10px13px15px16px25px
Shadow samples
none!important
0 0 35px 3px rgba(0,0,0,.33)
none!important
4px 0 rgba(0,0,0,.35) inset
rgba(82,82,82,.47) 1px 1px 3px
Character & component map
Shadow usage
heavy
Border usage
heavy
Gradients
present
Uppercase labels
yes (motif)
Letter-spacing
yes
Components observed vs gaps (gaps left empty — never auto-filled)
navobserved
heroobserved
buttonsobserved
cardobserved
pricingobserved
testimonialobserved
feature-gridobserved
galleryGAP
accordion-faqobserved
tabsobserved
tableobserved
formobserved
footerobserved
dropdownobserved
Worth considering — seen here, not yet captured
Notable styling present in this site's CSS that the tool doesn't yet turn into a captured value — candidates for future capture. Full coverage & roadmap →
filter (blur etc.)41×
object-fit30×
scroll-snap20×
mix-blend-mode17×
scroll-behavior7×
will-change (perf)2×
Beyond the capture — Claude's read & generative proposals
The character CSS mining can't reach — bespoke anatomy, motifs, the "feel". These are my interpretations and generative suggestions. They are NOT applied to the reconstruction above — they're ideas to react to when this becomes a library's seed.
Not reviewed yet. Ask Claude in chat to “read this teardown” and I'll add observations + generative proposals here.