Hey Product Hunt! 👋 I built RendShot because of a problem I kept running into while building AI automation workflows. I was automating content publishing to X, Xiaohongshu (RED), Instagram — the kind of stuff you wire up with n8n or OpenClaw. Every flow needed images: social cards, cover images, carousels. The default answer was "call DALL-E or Nano banana," but most of the time you don't actually need an AI image model. You need text, layout, and fonts rendered correctly. A $0.04 AI-generated image that takes 15 seconds and can't even render text properly is overkill when HTML can do the job in 2 seconds with pixel-perfect typography. So I built RendShot — send HTML/CSS, get back a PNG/JPEG/WebP. That's it. What makes it click in an AI agent workflow: 🤖 MCP server — Claude Desktop, Cursor, or any MCP-compatible agent can call RendShot as a tool. Your agent writes HTML, RendShot renders it. No prompt engineering for image generation, just deterministic output. 🧩 Templates — Design once in the visual editor, render variations by passing variables. Perfect for batch content: 50 social cards from a spreadsheet, daily quote images, product announcement posts. ⚡ 2-second renders — Fast enough to sit inside an n8n/OpenClaw flow without being the bottleneck. Fonts render perfectly every time — no more garbled text from AI image models. The whole stack is open source: TypeScript SDK, Python SDK, CLI, and MCP server. Free tier gives you 100 images/month. If you're building any kind of content automation pipeline, give it a try — would love to hear how you use it!

RendShot

HTML to Image API for developers and AI agents

HTML to Image API for developers and AI agents