their above-the-fold, as captured

UXXRAY · Teardown dossier ◉ scanned Jul 23, 2026

shunshi.ai

content · desktop · Roentgen 1.6

Biggest quick win

73% of measurable text passes WCAG AA. 4+ runs fail. Raise text/background contrast to ≥4.5:1 (≥3:1 for large text), especially on CTAs and body copy.

0 of 3 conversion signals detected

0 /100 coverage

Sparse ★☆☆☆☆ weighted for impact · 3 measured categories

UXXRAY scans for 3 conversion tactics with deterministic rules. Unconventional builds may not register, so this is detection coverage, not a verdict. Scored by Roentgen 1.6; detection improves, so rescan anytime.

Score breakdown

Action path 0/7
Friction reduction 0/6
Value reinforcement 0/7

The funnel, frame by frame

see their whole funnel without clicking through it. scored step by stepmeasured

Homepage / Lander

▲ strongest

0 tactics

no named tactics detected on this step

Action

the path to convert. CTA, sign-up, checkoutmeasured

0/7nonepoints earned

No signals detected for this category on the captured funnel.

Tactics we scan for but didn't detect

Add them to lift coverage, or tell us we missed one.

proof above the fold

Friction

objections answered, barriers removedmeasured

0/6nonepoints earned

No signals detected for this category on the captured funnel.

Tactics we scan for but didn't detect

Add them to lift coverage, or tell us we missed one.

FAQ / objection handling

Value

why it’s worth it. comparison and reinforcementmeasured

0/7nonepoints earned

No signals detected for this category on the captured funnel.

Tactics we scan for but didn't detect

Add them to lift coverage, or tell us we missed one.

us-vs-them comparison

UX signals

measured UX quality. contrast · mobile · type · a11y · layoutmeasured

Minor

2 flags · 5 checks

Measured UX quality. flags, not opinions. Deterministic, evidence-backed, and kept separate from the conversion score.

Readability & contrast

WCAG AA on measurable text. worst offenders quoted, never a fail we can’t compute.

NEEDS WORKText contrast (WCAG AA)

73% of measurable text passes WCAG AA. 4+ runs fail.

worst: “01” at 1.96:1 (24px, #b8aa92 on #f4ede0). needs 3:1

Raise text/background contrast to ≥4.5:1 (≥3:1 for large text), especially on CTAs and body copy.

Mobile & tap targets

Responsive viewport + touch-target sizes where it matters.

GOODMobile viewport

Responsive viewport meta is present.

Typography

Body size, line-height, line length. is the pitch actually readable?

NEEDS WORKBody text size

Smallest body text is 10px (base ~16px). below the ~16px readable floor.

Keep body copy ≥15-16px; reserve sub-13px for fine print only.

NOTELine length

1 paragraph runs past ~95 characters per line.

Cap measure at ~60-75ch for comfortable reading.

Accessibility hygiene

Language, alt text, form labels, named controls, headings, link clarity.

GOODLanguage declared

html[lang] is set for screen readers.

Next move

Now line it up against a competitor.

Paste any URL. We’ll run the same scan, the same rubric, and put the two funnels side by side. category by category, all measured.

Live scan · 15-25 sec · same rubric · nothing is guessed

Brand kit

palette by on-page usage · type rendered livemeasured

full palette ordered by on-page usage %

primary#f4ede0

accent#f4ede0

bg#faf6ed

surface#f4ede0

text#5b5147

accent

Body copy in their text role, on a surface card.

the roles, doing their job

The quick brown fox64px

Jumps over the lazy dog56px

A subheading in their voice48px

Body copy set at their measured paragraph size. This is how their product reads to a first-time visitor. same family, same scale, same weight.

List item one, at body scale
List item two
List item three

Teardown report. copy for your LLM

the full evidence-based read: what this funnel does well, where it is thin, and the on-page proof for each. paste into any agent to learn from

shunshi.ai · 0/3 tactics · 0 proof templates · evidence-backed

# Agent Brief. build in the style of shunshi.ai

> Machine-extracted from a live on-demand funnel teardown (2026-07-23). 100% facts —
> measured tokens, detected patterns, harvested proof strings. No LLM interpretation.
> Single-page capture (MVP): homepage/landing page only, no funnel walk.
> Tokens: `tokens.json` / `tailwind.css`.

**Category (label):** —
**Funnel:** Homepage / Lander

## Palette (area-weighted from live CSS)
| Hex | Usage | Likely role |
|-----|------:|-------------|
| `#faf6ed` | 58% | background |
| `#f4ede0` | 29.9% | background |
| `#f7f1e3` | 11% | background |
| `#2a2520` | 1.1% | support |
| `#9c6f2c` | 0% | accent/brand |

Curated roles: `primary=#f4ede0` · `accent=#f4ede0` · `bg=#faf6ed` · `surface=#f4ede0` · `text=#5b5147`

## Type
- Display: `Noto Serif TC`
- Body: `-apple-system`
- Size scale (px): 10, 11, 13, 14, 16, 19, 24, 36, 44, 48, 56, 64
- Radius (px): 8, 10, 999

## Patterns detected (fact-based). 0/3
- _(none detected)_

## Proof language (harvested strings → templates)
- _(none extracted)_

## Funnel facts
- Depth reached: 1 step(s) (single-page MVP capture)
- Proof points: 0 across 0 step(s)
- Discount/anchor pricing: no · Countdown: no · Subscription: no

## To replicate
Match the palette + type tokens above and deploy the detected patterns. This is a single-page capture. no cart/checkout tactics were evaluated.