Cortex Plugin: Canva / Image Generator

Generate social media graphics, presentations, logos, and diagrams from natural language using DALL-E, Stable Diffusion, and Canva.

Installation

cortex plugin install github:CortexPrism/cortex-plugin-canva

Tools

image_generate

Generate image from prompt.

• prompt (string, required) — Prompt describing the image to generate
• style (enum, default: "social") — Visual style: social, presentation, logo, diagram, photo, illustration
• size (string, default: "1024x1024") — Image dimensions
• count (number, default: 1) — Number of images to generate
• provider (enum, default: "dalle") — Generation provider: dalle, stable_diffusion, canva

image_edit

Edit existing image by providing an image URL and edit instructions.

• image_url (string, required) — URL of the image to edit
• prompt (string, required) — Edit instructions in natural language
• mask_url (string, optional) — URL of mask image

image_variations

Generate variations of an existing image.

• image_url (string, required) — URL of the source image
• count (number, default: 3) — Number of variations

canva_create_design

Create a Canva design from a template.

• template_type (enum, required) — Design type: instagram_post, facebook_post, presentation, logo, flyer, banner
• title (string, required) — Design title
• content (string, required) — JSON with text, images, and colors

canva_list_templates

List available Canva templates.

• category (string, optional) — Filter: social_media, presentations, marketing

Configuration

Key	Type	Description
openaiApiKey	secret	OpenAI API Key for DALL-E
canvaApiKey	secret	Canva API Key
stabilityApiKey	secret	Stability AI Key
defaultProvider	select	Default provider: dalle, stable_diffusion, canva
defaultSize	select	Default size: 1024x1024, 1792x1024, 1024x1792

Development

deno task test
deno fmt
deno lint

License

MIT