Vector Cover logo

Vector Cover

CommunityPopular
huytieu
vector-cover

Generate flat vector editorial cover illustrations in a warm editorial style: near-black canvas, cream organic blob, serif display type, one terracotta accent, and thin line-art concept motifs (grids, node charts, dashed "unknown" paths, mountains, question marks). Output is code (HTML + inline SVG) rendered to a crisp 2400x960 PNG via headless Chrome. Use when you want a blog, article or X-article cover image, a "hero graphic", or "make a cover for <title>".

Overview

Publisherhuytieu
RepositoryCOG-second-brain
Skill namevector-cover
Stars
1.3K
Forks
162
Bundled files
2
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.

  • 2 bundled files

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

  • Open source

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

Installation

Install the Vector Cover 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/huytieu/COG-second-brain/tree/main/.claude/skills/vector-cover
  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/huytieu/COG-second-brain.git /tmp/COG-second-brain
mkdir -p .claude/skills
cp -r /tmp/COG-second-brain/.claude/skills/vector-cover .claude/skills/vector-cover
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Vector Cover 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 Vector Cover 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 Vector Cover 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.

Vector cover

Configuration

The palette below is a default. Swap the hex values in template.html for your own brand, and keep the rule of exactly one accent color. Cover copy (eyebrow, title) comes from your article, not from this skill.

Requirements

  • Google Chrome or Chromium. render.sh finds it in the standard macOS locations or on PATH. Set CHROME to override.
  • No other dependencies, and no browser automation setup.

What this is

A designed vector graphic built as code: flat SVG on a small set of color tokens, rendered and screenshotted. It is not a diffusion image.

  • Yes: flat vector, geometric line-art, precise serif type, one accent color.
  • No: photorealism, gradients, glow, drop shadows, 3D, texture. If a request needs a photo, this is the wrong tool.

Design system (default palette, configurable)

TokenHexUse
ink#141413background canvas
ivory#FAF9F5the organic blob and primary headline text
clay#D97757accent: eyebrow, rule, the "unknown" motif
gridrgba(20,20,19,.09)faint grid drawn inside the blob
undercurve#E7D9C4faint warm curves near the blob's base

Rules:

  • Exactly one accent color (clay by default). Never add a second hue.
  • Serif for the eyebrow (italic) and the headline (bold). Sans only for tiny annotations.
  • Keep the dark field mostly empty. 3-5 decorative marks at most, low opacity.
  • Everything on the cream is clipped to the blob. Nothing floats half off it except the intentional motif that escapes (for example the dashed unknown path).

Canvas

1200 x 480 logical (2.5:1), rendered at 2x → 2400x960 PNG (X-article cover size). For a square social card use 1200x1200; for a 16:9 slide use 1280x720 and pass those to render.sh.

Layout skeleton (every cover)

+--------------------------------------------------------------+
|                                          ? (clay, top-right) |
|  A Field Guide:   (eyebrow, italic clay)   .-''''-. |
|                                                   /  BLOB   \|
|  Finding Your              (headline, bold ivory)|  + grid   |
|  Unknowns                                        |  + MOTIF  |
|  ___ (clay rule)                                  \        / |
|                                                    '-....-'  |
+--------------------------------------------------------------+
   left ~40% = type            right ~60% = blob + concept motif

The concept motif is the whole point

Every cover encodes the article's idea as a small line-art diagram inside the blob. Pick / adapt from this library (or invent one in the same visual grammar):

Article themeMotif
journey, growth, unknowns, "map vs territory"rising curve, solid→dashed, node dots, a ? and mountains at the frontier (the shipped template.html)
agents, systems, networks5-7 nodes joined by thin lines; one node filled clay = the active one
architecture, layers, stack3-4 stacked rounded rects, offset, one outlined clay
search, exploration, discoveryscattered dots + a dashed search path weaving between them + one magnifier ring
tradeoff, comparison, A vs Btwo curves crossing; the winner solid, the other dashed
pipeline, process, workflowleft→right stages (small circles) joined by arrows; last one clay
iteration, loops, refinementa spiral / concentric arcs tightening toward a center dot

Motif drawing rules: strokes #141413 at 2–2.6px, stroke-linecap="round"; hollow nodes = fill=ivory stroke=ink; filled nodes = fill=ink; anything "uncertain / future / unknown" is dashed and clay.

Workflow

  1. Get inputs from the user, or infer them from the article:
    • eyebrow (italic kicker, e.g. "A Field Guide:") - optional.
    • title (1–2 short lines; break it yourself for balance).
    • theme → pick a motif from the table above.
    • size (default 1200x480).
  2. Copy the template: start from template.html in this folder. It already has the tokens, blob, grid, type block, and the journey motif wired up.
  3. Edit three regions only:
    • the <text> nodes (eyebrow / two headline lines / rule width),
    • the CONCEPT MOTIF group (swap for the theme's motif),
    • the 3–5 decorative marks on the dark field. Leave the blob path, grid, and token colors alone unless deliberately restyling.
  4. Render:
    bash
    .claude/skills/vector-cover/render.sh <in.html> <out.png> [W] [H]
    # e.g. render.sh cover.html cover.png            -> 2400x960
    #      render.sh cover.html card.png 1200 1200    -> 2400x2400
    render.sh launches its own headless Chrome. It needs no remote-debugging port and no browser automation setup.
  5. Verify visually: Read the PNG. Check nothing is clipped off-canvas (the escaping motif must land on the dark field, not get cut), text fits the left ~40%, and the blob doesn't collide with the headline. Iterate on coords.
  6. Deliver the single PNG. Keep working files in your project folder.

Font note

The template uses a system serif stack (Charter/Iowan Old Style/Georgia). For a pixel-exact editorial look, drop a @font-face (e.g. a Tiempos/Copernicus- like serif) as a base64 data: URI at the top of the HTML - headless Chrome renders embedded fonts fine. Do not link remote fonts (offline render).

Gotchas

  • Off-canvas clipping: <text> anchored past x=1200 (or the escaping motif) gets cut. Keep the "unknown" cluster inside the canvas on the dark strip right of the blob (~x 1120–1195).
  • Headline overflow: at 72px the serif eats ~11 chars/line. Longer titles → drop to 60px or add a third line; re-check against the blob's left edge (~x498).
  • Blob edge vs grid: the grid is clipped to the blob, so extend grid lines well past the blob bounds - clipping handles the rest.
  • Never add gradients, shadows, blur, or a second accent color. One clay.

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 Vector Cover AI skill do?

Generate flat vector editorial cover illustrations in a warm editorial style: near-black canvas, cream organic blob, serif display type, one terracotta accent, and thin line-art concept motifs (grids, node charts, dashed "unknown" paths, mountains, question marks). Output is code (HTML + inline SVG) rendered to a crisp 2400x960 PNG via headless Chrome. Use when you want a blog, article or X-article cover image, a "hero graphic", or "make a cover for <title>".

Why use Vector Cover on TypingMind?

Because you install it once and use it with any model. Vector Cover 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 Vector Cover in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/huytieu/COG-second-brain/tree/main/.claude/skills/vector-cover. 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 Vector Cover?

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 Vector Cover?

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

Is the Vector Cover AI skill free?

Yes. It is published on GitHub by huytieu 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 👇