figma-design-compare
Skill4 starsChanged 27 days agoMITView on GitHub
Use when checking whether a built website, web app or mobile app screen matches its Figma design, doing design QA or a pixel-perfect review, comparing implementation with a Figma frame or link, or fixing "doesn't look like the design" feedback.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- gwair1989/figma-full-cycletheir own version, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- gwair1989/figma-full-cycle★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository30 other files it gives its agents.
- accessibility-auditSkill
- api-design-reviewSkill
- api-test-writerSkill
- architecture-diagramSkill
- bug-debuggerSkill
- bug-report-writerSkill
- changelog-writerSkill
- codebase-onboardingSkill
- code-reviewSkill
- database-schema-designSkill
- dependency-upgradeSkill
- design-handoff-specSkill
- design-reviewSkill
- dockerize-appSkill
- flaky-test-fixerSkill
- git-commit-writerSkill
- github-actions-ciSkill
- playwright-e2e-testsSkill
- prd-writerSkill
- proof-of-done-liteSkill
- readme-writerSkill
- refactor-plannerSkill
- security-auditSkill
- seo-auditSkill
- test-case-generatorSkill
- test-plan-writerSkill
- test-writerSkill
- ux-research-synthesisSkill
- ux-writingSkill
- web-performance-auditSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 23
Capture the build at the same size: Use the frame's width as the viewport width (for example 1440 desktop, 390 mobile) and the same scale.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Figma Design Compare
Overview
Compare what was built with what was designed using two kinds of evidence: screenshots side by side, and exact values from Figma (font, size, color, spacing, radius) against the values in code. Report every mismatch with the Figma value, the built value, where to fix it and how.
Inputs
- Figma frame link (
figma.com/design/<fileKey>/...?node-id=<id>) — one frame per screen. - The built screen: a URL or route, or a running emulator/simulator.
- Access to Figma: the Figma MCP server if connected, otherwise a
FIGMA_TOKEN(personal access token). Never print the token.
Method
- Get the design.
- Image of the frame at 2x and the node data (layout, text styles, fills, strokes, effects, auto-layout padding and gaps).
- Commands for MCP and REST API are in capture.md.
- Capture the build at the same size. Use the frame's width as the viewport width (for example 1440 desktop, 390 mobile) and the same scale. Load real fonts, wait for the page to settle, close cookie banners. Commands in capture.md.
- Visual pass. View both images. Walk the screen top to bottom, section by section: presence and order of elements, alignment, spacing rhythm, sizes, colors, typography, icons, images, borders, shadows, states.
- Measured pass. For each important element, read the value from the Figma node and from the implementation (source CSS/Tailwind/styles, or computed styles from the browser). Compare with the tolerances below.
- Locate the fix. Find the component and line that sets the wrong value.
- Report using the contract below. If asked, apply fixes and repeat steps 2-4 to confirm.
Tolerances
| Property | Match when |
| Spacing, size, position | within ±2 px at 1x |
| Font family, weight | exact |
| Font size, line height, letter spacing | within ±0.5 px |
| Color | same hex (or visually indistinguishable, ΔE < 2) including opacity |
| Border radius, border width | exact |
| Shadows | same offset, blur, color within tolerance |
Differences caused by real content length, OS font rendering or anti-aliasing are not mismatches.
Output contract
Start with: Screen: <name> — <N> mismatches (X high, Y medium, Z low). Match: <good | minor issues | off>.
Then a table, highest impact first:
| # | Element | Property | Figma | Built | Location | Fix |
| 1 | Hero heading | font-size | 56px | 48px | Hero.tsx:12 | text-[56px] |
Then Missing or extra elements, then Matches (what was checked and is correct). Impact: High = visible to every user at a glance (layout, missing element, wrong color or font); Medium = noticeable spacing or size; Low = within a few pixels.
Common mistakes
- Comparing at different viewport widths or scales, then reporting every size as wrong.
- Judging only from screenshots: a 4 px spacing error is invisible at a glance but obvious in values.
- Reading the wrong Figma variant or an outdated frame. Confirm the node id and last-modified date.
- Reporting differences in placeholder text or images as design defects.
- Fixing values with magic numbers when the project has design tokens for them.
Made by Wageesha Tripathi · LinkedIn · Email · More skills
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.