More screenshots
Works with the AI Agents you already use
Interface Quality Pass: Accessibility First, Polish Last
Run a priority-ordered design quality review over web, mobile, or desktop interfaces — accessibility and interaction first, polish last — with concrete checks per tier.
$5
Interface Quality Pass: Accessibility First, Polish Last
Example session with this skill installed
Review our new checkout page before launch. Tell us what would stop a real user, not what you would restyle.
- Read your context and instructions
- Compiled the interface quality pass
Ran the review in priority order, starting with accessibility. The checkout fails keyboard navigation at the payment step: the card-number field traps focus, so a keyboard user cannot reach the pay button at all — a blocker, not a preference. The error messages fail the next tier: an invalid card shows a red outline with no text, so a screen-reader user hears nothing and a colour-blind user sees nothing change. Interaction checks found a real bug: the pay button stays enabled while the request is in flight, so a double-tap submits the order twice — confirmed in testing with two identical orders created. Consistency checks found smaller items: the back link behaves differently on step two than on step three, and the success screen uses a different button style from the rest of the flow. Polish was checked last, and honestly: the spacing is uneven on the order summary, but nothing there stops a user, so it is listed as polish, not as a launch blocker. The deliverable is a ranked finding list: two blockers with the exact screen and the fix, three interaction and consistency items, and two polish notes — with the pre-ship gate stated plainly: the blockers must be fixed and re-tested before launch, and the rest can ship with tickets.
Connects securely to your tools. The creator never sees your data.
About this skill
Most interface reviews start with the colours and end before the keyboard ever gets tested. This skill runs a priority-ordered design quality review over web, mobile, or desktop interfaces — accessibility and interaction first, polish last — with concrete checks per tier and a pre-ship gate. It covers the review order that keeps a team honest: can everyone use it, does it behave correctly, is it consistent, and only then does it look good. It also covers the evidence: every finding names the screen, the check that failed, and the fix, so the review ends with work items instead of taste.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
It reviews what it can observe — it cannot test with real assistive-technology users, which is the stronger check.
A review is a snapshot: a fix can introduce a new issue, so blockers are re-tested after the fix, not assumed fixed.
Taste is not a blocker. The skill separates what stops a user from what a reviewer would restyle, and says which is which.
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 buying a skill.
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 3 days ago
Permissions
No special permissions declared or detected