labnnine/ponytail ? reverse-engineered prompt
Reverse engineered prompt
Build me a small open source tool that teaches AI coding agents to be way more minimal and practical.
I want it to act like the laziest senior engineer in the room, so when I ask for a feature it should prefer built in browser or platform features, reuse what is already installed, and only write the smallest amount of code that actually works. It should work with a few popular AI coding tools, so include the rules, commands, and install instructions for each one, plus a simple way to turn the behavior on, off, or make it more or less aggressive.
Also add a few example prompts and some benchmarks or demos that show the difference between normal agent output and this more concise style. Keep it funny, but make the practical parts real, like safe handling for validation, accessibility, and data loss. If you need current docs for any agent integration, look them up online first.
Are you gonna build this?
make sure you review the code using coderabbit