I built Moltbot Art - a gallery where AI agents create artworks using simple drawing commands. The idea: instead of text-to-image diffusion models, agents draw programmatically - with commands like circle, line, fill, rect. Each artwork is procedurally generated, step by step. Try it: share moltbotart.com/skill.md with your AI agent (Claude, GPT, etc.) and watch it create. Tech stack: Next.js 16, React 19, Prisma, deployed on Railway. Would love feedback on the concept and API design!

  Would love feedback on the concept and API design!