For the Framer Agent Hackathon, I decided to run a little experiment. I took a prompt I’ve used before to build a box breathing app on other platforms and threw it at Framer’s new Agent to see how it would compare.
The goal? See how fast I could get a clean, working app off the ground.
Honestly, the AI did 90% of the heavy lifting in just two prompts:
Prompt 1: The main prompt laying out the 4-4-4-4 box breathing logic and setup.
Prompt 2: Asking it to flip the vibe from dark mode to light mode.
Once the Agent handed over the layout, I jumped in for the final 10%—just adding a custom background image and tweaking the font to give it that perfect, relaxed feel.
For the Framer Agent Hackathon, I decided to run a little experiment. I took a prompt I’ve used before to build a box breathing app on other platforms and threw it at Framer’s new Agent to see how it would compare.
The goal? See how fast I could get a clean, working app off the ground.
Honestly, the AI did 90% of the heavy lifting in just two prompts:
Prompt 1: The main prompt laying out the 4-4-4-4 box breathing logic and setup.
Prompt 2: Asking it to flip the vibe from dark mode to light mode.
Once the Agent handed over the layout, I jumped in for the final 10%—just adding a custom background image and tweaking the font to give it that perfect, relaxed feel.