cowart-image-gen

Generates bitmaps with Vetta generate_image and inserts them into Cowart canvases via MCP.

154|27|Updated Aug 3, 2026
One-click install
npx skills add https://github.com/openvetta/open-vetta --skill cowart-image-gen-openvetta
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cowart-image-gen
Source: https://github.com/openvetta/open-vetta/tree/main/packages/plugins/externals/cowart-vetta/agent/skills/cowart-image-gen
Command: npx skills add https://github.com/openvetta/open-vetta --skill cowart-image-gen-openvetta

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Placing AI-generated images onto a Cowart canvas requires coordinating two separate toolchains: Vetta's image generation and Cowart's MCP canvas API. This Skill defines the exact workflow so images are generated at the right size and inserted or replaced in the correct location without manual file handling. ## Core Features & Use Cases - AI Image Holder Detection: Reads the current Cowart selection via get_cowart_selection and detects AI image holders to derive the target width and height for generation. - Bitmap Generation: Calls Vetta's generate_image tool with the holder's aspect ratio included in the prompt, then resolves the real local output path. - Canvas Insertion or Replacement: Uses Cowart MCP insert_cowart_image to either replace a selected AI image holder (replaceAiImageHolder: true) or insert the image onto the current page. - Use Case: A user selects an empty AI image placeholder on their Cowart canvas and asks for a product hero illustration; the Skill generates the bitmap at the holder's dimensions and swaps it into place. ## Quick Start Generate an image of a sunset over mountains and place it into the selected AI image holder on my Cowart canvas.

Frequently Asked Questions about cowart-image-gen

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an AI image onto a Cowart canvas?▼

Open the Cowart canvas, optionally select an AI image holder, then ask for the image. The Skill calls Vetta's generate_image to create the bitmap and uses Cowart MCP insert_cowart_image to place it on the current page or replace the selected holder.

How do I replace an AI image holder with a generated image?▼

Select exactly one AI image holder shape on the canvas before requesting generation. The Skill reads the holder's width and height from get_cowart_selection, includes the aspect ratio in the generation prompt, and calls insert_cowart_image with replaceAiImageHolder set to true.

What is the difference between generate_image and edit_image in Vetta?▼

generate_image creates a brand-new bitmap from a text prompt and is used by this Skill. edit_image revises an existing image or annotation screenshot and is handled by the separate cowart-image-edit skill instead.

Does this Skill work without the Cowart MCP plugin enabled?▼

No. The cowart-vetta plugin must be enabled so Cowart MCP tools like get_cowart_selection and insert_cowart_image are available. Opening the canvas first with open_cowart_canvas is also recommended.

Why does the generated image not match my holder's aspect ratio?▼

The target aspect ratio must be included in the generate_image prompt when replacing a holder. The Skill derives dimensions from the holder's props.w and props.h, so ensure the holder is selected before generation.