A small skill that makes AI responses more inspectable by requiring explicit confidence, remaining uncertainty, and hidden-assumption checks at the end of each answer.
This project was built as part of our June 2026 challenge: ship one small, judged AI-assisted product every day and be honest about where the tools help, where they get in the way, and where human taste still does the heavy lifting.
The score below reflects how strong the AI-assisted result is: not just whether it shipped, but whether it escaped generic slop and became something with a point.
Our rating
7.8/10
A solid result that mostly escapes slop. The idea is clear, the AI helped move it forward, and the remaining work is mostly product refinement.