unvibed.
Menu
Free UI audit

Agent-built demo / Before audit

Pourbook

E-commerce · Nov 8, 2026

Overall gradeFDesign grade after deductions

3 measured views. Local app. Design grades compare the actual screen with published product references. N/A means there was no applicable sample. Unverified means evidence was insufficient.

Product specificity and generic signals · F

Both captures show a concrete House filter recipe with 1:16 ratio, 240 g water and timed pours. The opening headline obscures the specific calculation task, and the coffee dose result is unreadable. (/)

Visual specificity before deductions C. Weighted design base C. Distinct signal deductions 6.

Product specificity supplies 55% of the base, and the five visual design dimensions share 45%. Round the base to the nearest letter. Each distinct verified generic signal lowers it by one letter, down to F. Foundations are reported separately. The detector ceiling is F. Vision reviewed 1 of 1 measured routes.

Template copy · −1
“all-in-one solution” does not identify a product task or outcome. (/, 1440 px, Detector and vision.)
Gradient hero · −1
A gradient fills the opening section around the visible heading. (/, 1440 px, Detector and vision.)
Emoji bullets · −1
3 visible list items or short passages start with emoji. (/, 768 px, Detector.)
Default shadcn grays · −1
3 separate visible panels repeat the same near-default neutral border, fill, width and corner treatment. (/, 768 px, Detector.)
Purple on white · −1
A prominent purple action or heading sits on a white surface. (/, 1440 px, Detector and vision.)
Centered everything · −1
The heading, 6 visible text blocks and a primary control share centered alignment. (/, 768 px, Detector and vision.)

Compared with the 2026 references

  • / · The reference makes its object and purchase decision explicit through product views, finish choices, price and stock. Pourbook supplies useful recipe proof but leaves the core dose result unreadable. Commercial grade: F after the four observed signal deductions; specificity carries 55%, with visual foundations sharing 45%. Spree demo · automatic espresso machine

Visual design

Hierarchy
C
At 1440, the oversized hero dominates the recipe; at 390, it pushes the recipe below substantial promotional content. The useful dose result receives less emphasis than Get started. (/)
Typography
B
At both widths, bold recipe naming, aligned quantities and smaller metadata support scanning. The phone headline wraps cleanly. The dose area lacks readable numeric text. (/)
Spacing
B
At 1440, the recipe card has comfortable internal spacing. At 390, margins narrow while the chart, schedule and footer remain contained and distinct. (/)
Copy and labels
C
House filter, Bloom, First pour and Save recipe are specific at both widths. Transform your coffee experience, ultimate all-in-one solution and Get started fail to explain dose calculation or the next destination. (/)
Imagery
N/A
This recipe task requires no photograph or illustration. Both widths provide a labeled pour-profile chart as functional product information.

Foundations · B

Contrast, responsive layout and keyboard reach are sampled separately. The foundations grade is the lowest verified result in this group.

Color contrast
B
Checked 10 solid-background text samples across the captured views. No issue detected by this check.
Responsive layout
A
All sampled viewports fit without horizontal overflow and contained readable text. Verified across 3 measured views.
Keyboard access
A
Tab reached two visibly focused controls and Shift+Tab returned to the first. Verified across 3 measured views.

Fix these first

  1. Name the object, action and result in the heading.Product specificity · / · 1440 px
  2. Replace the opening gradient with a view of the product doing its work.Product specificity · / · 1440 px
  3. Use readable labels and group the steps around their actual task.Product specificity · / · 768 px
  4. Give the primary task, supporting information and actions distinct visual roles.Product specificity · / · 768 px
  5. Choose an action color grounded in the product and separate it from status colors.Product specificity · / · 1440 px

Evidence from the screens

Gray blocks conceal form values or sensitive text. Numbered outlines identify measured issues.

Captured / at 390 pixels wide
/ · 390 × 844
  1. “all-in-one solution” does not identify a product task or outcome.
Captured / at 1440 pixels wide
/ · 1440 × 900
  1. A prominent purple action or heading sits on a white surface.
  2. A gradient fills the opening section around the visible heading.
  3. “all-in-one solution” does not identify a product task or outcome.
Captured / at 768 pixels wide
/ · 768 × 1024
  1. The heading, 6 visible text blocks and a primary control share centered alignment.
  2. 3 separate visible panels repeat the same near-default neutral border, fill, width and corner treatment.
  3. 3 visible list items or short passages start with emoji.
  4. “all-in-one solution” does not identify a product task or outcome.

Independent task review

The reviewer assesses each route against its user task using desktop and phone evidence. Its task grades describe the interaction evidence available for that route.

/

Audit Pourbook’s own landing interface at /: calculate a coffee dose and find the House filter pour-over recipe, then save it.

Reviewed at 1440 × 900 and 390 × 844.

Task hierarchy
C
At 1440, the large promotional hero precedes the useful House filter recipe. At 390, the recipe remains readable, but the Coffee dose area shows no legible numeric result. Get started competes with the task-specific Save recipe action at both widths.
Landing message
C
Both widths identify coffee brewing through the recipe, ratio and pour schedule. The headline and supporting copy describe an interchangeable improvement rather than dose calculation or recipe management. Get started does not explain its destination.
Interaction feedback
Unverified
The supplied checks record visible controls and a reached focus target at both widths. They contain no dose calculation trigger, result, save confirmation or persistence outcome.
Access and recovery
C
At 390, navigation, chart, recipe rows and Save recipe fit without visible horizontal clipping; at 1440 they are legible. Coffee dose has no readable value at either width. One visible Tab target was reached per width; complete keyboard operation and recovery were not exercised.
Release evidence
Unverified
The local captures establish the opening layout at 1440 × 900 and 390 × 844. Checks report one primary heading and six visible controls at each width. Calculation accuracy, saving, reload persistence, error recovery and deployed behavior remain unobserved.

First action

Make the House filter Coffee dose value readable at both widths, with its unit and relationship to 240 g water at 1:16 clearly stated.

How to verify At 1440 and 390, verify that the recipe displays 15 g coffee for 240 g water at 1:16. Exercise any dose control by keyboard and confirm its result updates. Separately test Save recipe and its confirmation. Contrast measurements and complete keyboard reach remain unverified; responsive fit is observed only at the two supplied sizes.

What a repair could look like

Second Cast is an agent-built used tools marketplace demo. Both captures show the same task.

Second Cast before the rebuild
Before
Second Cast after the rebuild
After
Compare Second Cast at both widths →