- Home
- Skills
- Testing & Debugging
- ScrollProof
More screenshots
Works with the AI Agents you already use
ScrollProof
Audit GSAP, ScrollTrigger, and reveal animations for breakage, layout shifts, and performance jank.
$2
ScrollProof
Example session with this skill installed
Audit my local Vite site at http://localhost:5173 for scroll jank and check why the hero section isn't pinning correctly.
- Read your context and instructions
- Compiled the scrollproof
- Generated the document
I have completed the scroll-driven animation audit for your local dev server.
Summary: Score 74/100. Found 2 errors and 3 warnings. The hero section fails to pin because the end trigger is unreachable.
Animation Audit Report
| # | Severity | Scroll | Element | Issue | Source |
|---|---|---|---|---|---|
| 1 | ERROR | 0px | section.hero | Broken pin: trigger end precedes start | /src/components/Hero.tsx:42 |
| 2 | ERROR | 1450px | div.reveal-box | Content flash: opacity 1 before intro | /src/styles/main.css:112 |
| 3 | WARN | 800px | img.parallax | Layout shift: missing height/width | index.html:22 |
| 4 | WARN | 2200px | nav.sticky | Forced reflow during scroll event | /src/hooks/useScroll.js:15 |
| 5 | WARN | - | GSAP Engine | 2 stale ScrollTriggers detected | N/A |
Technical Analysis
The error in Hero.tsx at line 42 is caused by a ScrollTrigger where the end value evaluates to a position above the start position. This usually happens when the trigger element has no height or the end marker is calculated before images load. In main.css, the reveal-box lacks an initial visibility: hidden or opacity: 0 in the base CSS, causing it to flicker before GSAP takes control.
The performance trace shows significant jank (dropping to 42fps) at the 2200px mark due to a scroll listener in useScroll.js triggering a layout-heavy getBoundingClientRect() call.
Next steps
- Add
opacity: 0to.reveal-boxinsrc/styles/main.cssto prevent FOUC. - Review the
endcalculation inHero.tsxline 42; ensure the trigger element has a defined height. - Open
.scrollproof/localhost-5173/report.htmlin your browser to view the frame-time chart.
Hero.tsx
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Scroll-driven animations often fail silently, causing layout shifts, broken pins, or elements that never reveal. This skill objectively audits your site in a headless browser to identify exactly where and why GSAP, ScrollTrigger, or native scroll effects break.
What it does
- Automated scrolling simulates user behavior across different viewport sizes and CPU speeds.
- State capture records every animated element's properties at specific scroll intervals.
- Performance tracing detects jank, forced reflows, and layout shifts during scroll.
- Engine inspection probes GSAP and ScrollTrigger instances for stale or unreachable triggers.
- Visual auditing captures annotated screenshots of broken pins and content flashes.
How it works
- Environment check verifies the presence of Node.js and a Chromium-based engine.
- Audit execution runs the CLI against a local dev server or live URL with optional throttling.
- Data collection generates a performance trace and a detailed report in a local directory.
- Issue reporting summarizes errors by severity, scroll position, and source file line.
Frameworks & tools
This tool targets sites using GSAP, ScrollTrigger, Lenis, and standard CSS reveal-on-scroll effects. It requires Node.js 18.17+ and Playwright Chromium.
Why this beats prompting it yourself
Generic LLM prompts cannot see "invisible" animation states or feel frame-rate drops. This skill uses a real browser engine to track the DOM at specific scroll offsets, catching logic errors that don't trigger console logs.
Use cases
- Debugging GSAP ScrollTrigger pins that jump or overlap other content.
- Identifying "invisible" sections caused by opacity tweens that never trigger.
- Stress-testing scroll performance on simulated mid-range mobile devices.
- Validating that reveal animations work correctly after a framework migration.
Known limitations
Requires a Chromium-based browser installed on the host. Does not support Safari-only WebKit bugs or Firefox-specific rendering issues.
Requires local Chromium/Chrome. Only audits scroll-driven logic; cannot fix high-resolution asset loading delays. Best used on local dev servers providing source maps.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 19 days ago
Requires