Hey everyone, I’m Peter a designer and developer. I made this new tool i call Design In The Browser. I built this because I kept running into the same problem: explaining visual changes. I’d end up constantly taking screenshots, copying them into the terminal, and writing long prompts describing which element, where it is, what to change, then repeating when the AI guessed wrong, when all I really wanted to do was point at it.
Design In The Browser lets you click any element on your page and send it directly to Claude Code, Cursor, or Gemini CLI with full context. The AI knows exactly what you’re looking at, so you skip the back-and-forth. It also has an integrated terminal, viewport switcher for responsive testing. A cool thing is that you can also queue prompts while Claude Code is working and batch multiple edits.
It’s macOS only for now, free to use, and works with any local dev server. Would love feedback, especially on what features you’d want next.
Hey everyone, I’m Peter a designer and developer. I made this new tool i call Design In The Browser. I built this because I kept running into the same problem: explaining visual changes. I’d end up constantly taking screenshots, copying them into the terminal, and writing long prompts describing which element, where it is, what to change, then repeating when the AI guessed wrong, when all I really wanted to do was point at it.
Design In The Browser lets you click any element on your page and send it directly to Claude Code, Cursor, or Gemini CLI with full context. The AI knows exactly what you’re looking at, so you skip the back-and-forth. It also has an integrated terminal, viewport switcher for responsive testing. A cool thing is that you can also queue prompts while Claude Code is working and batch multiple edits.
It’s macOS only for now, free to use, and works with any local dev server. Would love feedback, especially on what features you’d want next.