gocko1004/everything-claude-code ? reverse-engineered prompt

Build me a cross platform AI agent setup that makes Claude Code, Codex, Cursor, and similar tools work better out of the box.

I want a full system, not just a few config files. It should include reusable skills, memory that carries across sessions, hooks for start and stop events, safe security checks, and commands that help the agent stay focused and research first. Make it easy to install, update, and turn on only the parts I need. If possible, include support for different languages and workflows so it can help with real projects, reviews, build issues, documentation lookup, and general productivity.

Please make it feel production ready, with clear docs, sensible defaults, and compatibility across the main agent tools people actually use. If you need to look up current docs online for any of the agent platforms, do that.