Coordinate canvas projects, AI image generation, and design-marker edits.
npx skills add https://github.com/dtyq/magic --skill canvas-designer
Please help me install this Agent Skill. Skill: canvas-designer Source: https://github.com/dtyq/magic/tree/main/backend/super-magic/agents/skills/canvas-designer Command: npx skills add https://github.com/dtyq/magic --skill canvas-designer
This skill consolidates project planning, multimedia principles, AI image generation, web image search, and design-marker processing into a unified canvas design workflow. It enables teams to plan, create, and edit canvas projects without switching tools.
Load the canvas-designer skill and start a new project to begin creating a canvas using AI-generated images and marker-driven edits.
High-intent search queries and answers about installing and using this skill.
To design canvases with AI image generation, load the canvas-designer skill to orchestrate project creation, generate visual assets, and apply design markers to specific image regions while preserving the rest of the canvas.
Yes, the unified canvas workflow integrates web image search directly, allowing you to populate canvases with relevant visuals and reference material without switching between separate design tools.
Design markers parse user intent and apply edits to specific regions of an image, enabling precise multimodal asset edits across iterations while preserving the rest of the canvas.
Yes, the canvas workflow extends to video generation and animation tasks using dedicated reference docs, allowing you to build out dynamic visual content within the same project environment.
The best way to manage multimodal assets is by loading the unified canvas workflow skill, which coordinates project planning, AI image generation, and design-marker processing across all editing iterations.
TechNickAIPerform web research and document findings in research/[topic].md.
liamtran96Analyze and generate multimedia content using Google Gemini multimodal APIs.
jmaglyPackage validated skill directories into ZIP files for Claude AI uploads.
the-open-agentEdit specific PDF pages using natural-language instructions via the nano-pdf CLI.
