This handles the full loop of creating banners with AI image generation: you discuss what you need, it generates 20 variations at once using the nanobanana skill, shows you an HTML preview, then iterates based on which ones you like. Once you pick a winner, it crops to your target platform ratio (GitHub README 2:1, Twitter header 3:1, etc). The workflow is thorough, maybe even a bit prescriptive, but that's helpful when you're staring at a blank canvas. It saves everything to a dated archive folder and knows common banner sizes off the top of its head. Good for when you need a header image and don't want to fire up Figma.
npx -y skills add resciencelab/opc-skills --skill banner-creator --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
cursor/plugins
github/awesome-copilot
alirezarezvani/claude-skills
microsoft/win-dev-skills