Web Visuals logo

Web Visuals

Organization
glifxyz
web-visuals

Make images, textures, illustrations and background video for a website or app you are building, with Glif, and wire them into the code. Use when: "add a hero image", "this page needs visuals", "replace these stock photos", "placeholder images", "background texture", "seamless pattern", "paper grain", "section backgrounds", "feature illustrations", "SVG illustration", "empty state art", "404 illustration", "infographic", "blog cover", "OG image", "social share image", "product cutouts for the shop", "looping background video", "scroll animation video", or when a page you're building has empty image slots. NOT for: editing an existing image (use edit-images), images not tied to a site (use make-images), logos or icons (use app-icons-and-logos), store or launch graphics (use app-promo-materials), demo videos (use app-demo-videos), game art (use game-assets).

Overview

Publisherglifxyz
Repositoryglif-mcp-server
Skill nameweb-visuals
Stars
215
Forks
32
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by glifxyz on GitHub. Read the source before you install it.

Installation

Install the Web Visuals AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/glifxyz/glif-mcp-server/tree/main/skills/web-visuals
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/glifxyz/glif-mcp-server.git /tmp/glif-mcp-server
mkdir -p .claude/skills
cp -r /tmp/glif-mcp-server/skills/web-visuals .claude/skills/web-visuals
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Web Visuals in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Web Visuals on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Web Visuals is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Make the visuals a web page needs and put them in the project. The user's explicit instructions come first, within host safety rules.

1. Take inventory

Read the page or component. List each visual slot:

  • Hero: full-bleed or split, and which side the headline sits on.
  • Section backgrounds: soft gradients, paper grain, material close-ups.
  • Textures and patterns: must tile seamlessly.
  • Illustrations: feature cards, empty states, 404, onboarding. Ask for SVG when the style is flat.
  • Cutouts: product or people photos on a transparent background, to sit on page colors.
  • Infographics and stat cards: keep the numbers and labels in HTML; let Glif make the art around them.
  • Device or hardware surface: a photoreal laptop, phone or panel with a pure #00FF00 screen area that your code draws live UI into.
  • OG image: 1200×630, one per major route.
  • Motion (optional): a looping background, or a scroll-scrubbed hero video.

Use what the user already has first. Never replace a supplied logo, product photo or screenshot. Glif fills the gaps.

2. Match the site's style

Pull the palette (CSS variables or Tailwind theme), fonts, mood and any existing images. Put hex colors and a one-line mood in the brief. If a logo or screenshot sets the style, upload it with upload_file and attach the returned fileUrl.

If the site supports dark mode, ask for matched light and dark versions with the same size and crop.

3. Check cost, then batch

Generation spends the user's Glif credits.

  • If the user asked for these visuals, go ahead.
  • If adding visuals was your idea, show the slot list in one line each and wait for a yes.
  • Video uses more credits than images. Offer it; never make it unasked.
  • When the look is still open, ask for 2 or 3 directions for the hero first, then make the rest in the chosen style.

Send one compose_project call for the whole page. For each asset, give:

  • filename, such as hero.png or texture-paper.png
  • exact pixel size and aspect ratio
  • where it goes and what it must show
  • where to leave empty space for headline text

Add these rules to every brief:

  • No text, letters, logos or watermarks in images, except an OG card that should carry the page title. Real text stays in HTML.
  • Textures: seamless on all four edges, 1:1, low contrast so text on top stays readable.
  • Report each output's actual pixel size.

Keep coding with placeholders while it runs. Call get_job_status with the jobId as job_id when you need the results. A pending status or a timeout is not a reason to start another run. Share the projectUrl.

4. Bring the files home

Never hotlink Glif URLs in shipped code. Download each result into the static folder, such as public/images/ (Next.js, Vite, Astro) or static/ (SvelteKit):

sh
curl -fsSL "<media url>" -o public/images/hero.png
file public/images/hero.png
  • Use the media's MIME type for the extension. Don't overwrite an existing file; save hero-v2.png instead.
  • Convert photos to WebP or AVIF if a tool is available, such as sharp, cwebp, magick or sips. Aim for about 300 KB for a hero, much less for other images.
  • Check each texture by tiling it 2×2. If a seam shows, ask Glif to fix that one in the same project.
  • Check real pixel sizes. Resize locally or ask Glif to redo a wrong one. Never upscale.
  • Check any text in an OG card letter by letter.

5. Wire it in

  • Set width and height (or use the framework's image component) to prevent layout shift.
  • Hero: fetchpriority="high", no lazy loading. Everything else: loading="lazy".
  • Add srcset widths such as 640, 1280 and 1920 when the framework doesn't.
  • Write real alt text. Use alt="" for decorative textures.
  • Textures: background-repeat: repeat with a sensible background-size. Grain overlays work well at 4–10% opacity.
  • Background video: <video autoplay muted loop playsinline poster="...">, MP4 (H.264), no audio track.
  • OG image: absolute URL in og:image and twitter:card set to summary_large_image.

Scroll-scrubbed hero video. Ask Glif for 3 to 5 keyframe stills first. After the user approves them, ask for smooth transitions between them, stitched into one MP4 encoded so every frame is a keyframe, for smooth seeking. In code, pause the video and set video.currentTime = progress * video.duration on scroll, inside requestAnimationFrame. Show the poster image when the user prefers reduced motion.

Then render the page, if you can, at narrow and wide widths, in light and dark mode. Report what you added and where. Report anything you couldn't check.

When not to use Glif

Simple shapes, gradients, dividers and standard UI icons are better as CSS, hand-written SVG or the project's icon library. Don't spend credits on them.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Web Visuals AI skill do?

Make images, textures, illustrations and background video for a website or app you are building, with Glif, and wire them into the code. Use when: "add a hero image", "this page needs visuals", "replace these stock photos", "placeholder images", "background texture", "seamless pattern", "paper grain", "section backgrounds", "feature illustrations", "SVG illustration", "empty state art", "404 illustration", "infographic", "blog cover", "OG image", "social share image", "product cutouts for the shop", "looping background video", "scroll animation video", or when a page you're building has emp...

Why use Web Visuals on TypingMind?

Because you install it once and use it with any model. Web Visuals is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Web Visuals in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/glifxyz/glif-mcp-server/tree/main/skills/web-visuals. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Web Visuals?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Web Visuals?

As many as you like. As long as a model supports skills, you can use Web Visuals with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Web Visuals AI skill free?

Yes. It is published on GitHub by glifxyz under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇