hey all, I'm Nate. By day, I'm working at an ai native media agency solution. By night, I (ie me and my agents) build small apps on Vercel.
While building, the gap I kept hitting was having something break only for me to find out later and push a fix hours / days later. Sure, the fix was easy, shared with my agent, but it should be automated.
Spooky Particles closes that loop. It checks every 5 mins. When something breaks, it pings your coding agent the cause. Your agent fixes, tests, and ships the fix (or opens a PR, whatever your rules are). Spooky then checks again, and "fixed" means the check passes. It also notifies you via email or ntfy.
hey all, I'm Nate. By day, I'm working at an ai native media agency solution. By night, I (ie me and my agents) build small apps on Vercel.
While building, the gap I kept hitting was having something break only for me to find out later and push a fix hours / days later. Sure, the fix was easy, shared with my agent, but it should be automated.
Spooky Particles closes that loop. It checks every 5 mins. When something breaks, it pings your coding agent the cause. Your agent fixes, tests, and ships the fix (or opens a PR, whatever your rules are). Spooky then checks again, and "fixed" means the check passes. It also notifies you via email or ntfy.
Setup is one prompt you give your agent: https://spookyparticles.com/setup.md
Feedback would be appreciated, looking for 2 things: where setup is rough, and where you would not trust an agent to ship a fix without you.
Thanks all!