CodeWithJuber/forgekit ? reverse-engineered prompt
Reverse engineered prompt
Build me a tool called Forge that gives my AI coding assistants one shared brain.
I want to write my rules, memory, and safety notes once, then have it generate the right config for tools like Claude Code, Codex, Cursor, Gemini, Aider, and the other common ones. It should keep a team memory of useful lessons and decisions, suggest what files might be affected before a change, and add guardrails so the agents remember important project rules even across sessions.
Make it feel like a simple command line app that can initialize a repo, check whether everything is wired up, and sync shared memory back into git friendly files. It should be opinionated but easy to use, with clear docs and a smooth setup path. If you need to look up current docs for any agent integrations, go ahead and do that.
Are you gonna build this?
make sure you review the code using coderabbit