LaunchKit · 2026
Back to Skills

canvas-design

Create beautiful visual art in .png and .pdf documents.

0
0 downloads
by @seanphan

Setup & Installation

openclaw skills install @seanphan/canvas-design

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install canvas-design

What This Skill Does

Generates original visual art as PNG or PDF files using a two-step process: first defining a named design philosophy (an aesthetic movement with principles of form, color, and composition), then rendering that philosophy as a canvas artwork. Output is predominantly visual, with any text used sparingly as a design element rather than content.

Separating philosophy creation from visual execution lets each stage be refined independently, which reduces the common AI output problem of visually generic or compositionally unconsidered results.

When to use it

  • Making a custom poster for an event or occasion
  • Generating a printable art piece for a home or office wall
  • Producing a visual concept sheet for a brand or project pitch
  • Creating a cover image for a zine, booklet, or publication
  • Designing an abstract art print as a personalized gift

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: Create a poster inspired by late Soviet brutalism

AGENT
  1. 1Names a design movement and writes a 4-6 paragraph visual philosophy as a .md file covering form, color, scale, and composition
  2. 2Identifies a subtle conceptual thread from the user's request to embed as the artwork's underlying reference
  3. 3Selects fonts from the ./canvas-fonts directory that match the philosophy's tone
  4. 4Renders the canvas with repeating geometric patterns, a limited color palette, and minimal integrated typography
  5. 5Refines composition for cohesion and precision, then outputs the final .png or .pdf alongside the .md philosophy file
OUTPUT

A .md design philosophy file and a .png or .pdf canvas artwork with museum-quality visual execution