LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|
Skillofficial publisher

web-design-guidelines (Vercel)

Skill : A reusable method: instructions, procedures, sometimes scripts or templates · requires claude-code

Audits your interface code against more than one hundred Vercel rules (accessibility, forms, animations, images) and reports gaps file by file.

Excellent UI reviewer after generation. It downloads its rules at each run from a Vercel repository, hence the orange badge.

Overall score
8.1 /10
Security: one caveat, see the card
Utility
7.0
Reliability
10.0
Security
One caveat
Ease
1 click · Beginner

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

When generating UI code with an AI agent, it is hard to know whether the result follows web best practices: accessibility, focus management, forms, animations, typography, images, performance, internationalisation. web-design-guidelines fills that gap by automatically auditing the produced code against more than one hundred rules.

In practice, the skill is installed inside claude-code and is triggered by phrases such as "Review my UI" or "Check accessibility". It reads the project's interface files, downloads the reference rules from the Vercel repository on GitHub at each run, then reports detected gaps file by file, grouped by category.

It is aimed at web developers who already use claude-code and want a quality safety net after a code generation phase. It covers eleven categories: accessibility, focus states, forms, animations, typography, images, performance, navigation and state, dark mode, touch interactions, and internationalisation.

It is not suited to those looking for an automated testing tool integrated into a CI/CD pipeline, nor to those who do not use claude-code as their development environment.

What it offers

  • ·Accessibility audit: checks aria-labels, semantic HTML, and keyboard handlers.
  • ·Focus state review: detects missing or incorrect focus-visible patterns.
  • ·Form verification: checks autocomplete, validation, and error handling.
  • ·Animation analysis: flags missing prefers-reduced-motion support and non-compositor-friendly transforms.
  • ·Image audit: checks dimensions, lazy loading, and alt text.
  • ·Typography check: detects straight quotes, incorrect ellipsis characters, and missing tabular-nums.
  • ·Internationalisation check: verifies use of Intl.DateTimeFormat and Intl.NumberFormat.
  • ·File-by-file report: returns detected gaps grouped by category for each audited file.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
GitHub vercel-labs/agent-skills, skills.sh
Host
claude-code
Licence
MIT
Other hosts
claude-code, cursor, codex, opencode, gemini-cli
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Your data
Data stays local
Maintenance
Active
Last release
28 August 2026

Links and repository checked on 25 October 2026 · added on 23 September 2026

How to install web-design-guidelines (Vercel)

1. Make sure Node.js is installed and claude-code is configured on your machine. 2. Open a terminal at the root of your project. 3. Run the installation command: npx skills add vercel-labs/agent-skills 4. The web-design-guidelines skill is included in the collection and becomes automatically available in claude-code. 5. In claude-code, type a phrase such as "Review my UI" or "Check accessibility" to start an audit. 6. Note that each audit downloads rules from GitHub: occasionally check their content if you want to know exactly what is being applied.
npx skills add vercel-labs/agent-skills

Official documentation ↗

In the same category

Whole category Design & slides →