Publisher: Vercel · Editorial profile by AI Tools Ratings · Review method
Not runtime tested; source-document review only. No security certification.
What task is this skill designed for?
Inspect a form component for missing labels, unclear focus states and other applicable interface rules, then prioritize issues that can be verified in the browser.
What input and output should you expect?
Input: A specific file or glob, the intended interface behavior and the code revision under review.
Output: Expected, not independently tested: a concise file-and-line issue list based on the fetched guidelines, followed by any separately authorized fixes.
What does it need to run?
An agent capable of reading local files and fetching the guideline URL referenced by the source. Browser testing is still needed for keyboard behavior, actual contrast and responsive interaction.
The author repository publishes Agent Skills-format instructions. Client-specific installation and tool dependencies must be checked; no Codex, Claude Code or Cursor runtime compatibility is independently verified here.
How should you approach installation?
- Inspect the pinned source, including the external guideline URL it will fetch.
- Use the repository’s current installation guidance and confirm the agent can retrieve references safely.
- Trial a small non-sensitive component and save the guideline version alongside the review.
Installation guide: discovery, dependencies and runtime checks
What files, accounts or data can it access?
Reads project source and makes an external request for the current guidelines. The source points to a moving main-branch document; record its version when reproducibility matters and treat its contents as untrusted reference data.
What does the skill cost?
The repository README declares MIT licensing. Agent usage and separate audit tools may add costs.
What are its limits and alternatives?
Pinning SKILL.md does not freeze the remote guideline file it fetches. A static review is not a complete accessibility audit, and a file-and-line report can miss browser-only behavior or backend validation problems.
Use manual keyboard checks and appropriate accessibility testing tools to verify actual behavior. Choose this skill for an initial, bounded source review with concrete code locations.
What could you ask in a small trial?
Try this with a small sample:
Review only the signup component against the fetched interface guidelines. Record the guideline revision, report file-and-line evidence, and separate source findings from browser checks that still need execution.
How would you verify the result?
- Confirm which remote guideline content and version were used.
- Open each reported location and check that the rule applies.
- Verify keyboard focus, labels and error feedback in the rendered interface before closing an issue.
Use a baseline and record acceptance evidence before treating a successful task as a reliable workflow.
Which source and license were checked?
Read the reviewed SKILL.md at its pinned commit · Complete reviewed folder
Repository and current publisher guidance
Checked:
MIT — declared in the repository README at the reviewed commit; no separate license file was found in the inspected root or skill folder. Confirm third-party material terms before redistribution. License evidence
Version details and review scope
Source: vercel-labs/agent-skills/skills/web-design-guidelines
Reviewed commit: 063bee94c3f4df8453406c830b0a7df0f2860278
SKILL.md SHA-256: f4647ca866a3accf763777f83e7682954f0187cd6bea7eea0399796652414e8f
Source-document review. Source text and stated dependencies were inspected for this profile. Referenced code was not comprehensively audited, no runtime test was performed, and source files are not mirrored here.