Nasiko-Labs/nasiko ? reverse-engineered prompt
Reverse engineered prompt
Build me a local first control plane for AI coding agents that can find what is already running on a developer machine, show each agent in one place, and let me see what it is doing, what model it is using, and roughly what it costs.
I want a dashboard plus a command line tool that can discover supported agents like Claude Code, Codex, OpenCode, and Cursor, then let me opt into reporting and routing without changing how people normally use those tools. It should be able to install and remove hooks, sync session events safely even if the server is temporarily down, and keep the history on my side.
Also add a way to route model calls, manage auth, and show traces, token usage, and spend in the UI. If you need to look up current docs online, go ahead. Keep it practical and make it feel like an open runtime I can run with Docker or from source.
Are you gonna build this?
make sure you review the code using coderabbit