Launching today
ScribUI
A visual tool for reviewing AI-generated interfaces
1 follower
A visual tool for reviewing AI-generated interfaces
1 follower
Point, don't describe. ScribUI turns visual feedback on your app into precise instructions for your coding agent. Works with Claude Code, Codex, Cursor and any agent that reads files.
Free
Launch Team
Render Sandboxes: a computer for every agentRun untrusted code safely. Idle sandboxes cost nothing
Promoted
As a Design Engineer and Frontend Developer, I've always relied on a tight feedback loop. Change some code, refresh the browser, see how it looks and feels, and keep refining until it feels right. With AI coding agents, I'm getting things implemented much faster. But I'm also losing some of that granular control. There's suddenly a lot of prompting back and forth over small details.
"𝘈𝘥𝘥 5𝘱𝘹 𝘮𝘰𝘳𝘦 𝘱𝘢𝘥𝘥𝘪𝘯𝘨 𝘵𝘰 𝘵𝘩𝘦 𝘧𝘭𝘰𝘢𝘵𝘪𝘯𝘨 𝘮𝘦𝘯𝘶."
"𝘔𝘢𝘬𝘦 𝘵𝘩𝘦 𝘭𝘰𝘨𝘰 20% 𝘭𝘢𝘳𝘨𝘦𝘳."
Things I used to tweak in seconds now require explaining what looks wrong and hoping the agent understands. Figma MCP helps a lot with implementing designs correctly. But getting the feel right is still difficult. Some things are just hard to describe in words.
𝗦𝗼 𝗜 𝗯𝘂𝗶𝗹𝘁 𝗦𝗰𝗿𝗶𝗯𝗨𝗜.
A lightweight visual review layer between you and your AI coding agent. Not a replacement for Figma, and not another AI tool.
The idea is simple: browse your running application, point at something that doesn't feel right, draw or comment directly on it, and let your agent handle the changes.
It works alongside your existing agentic setup and runs in the browser or as a native app on macOS, Windows, and Linux, with a built-in application preview.