I built a state-invariant execution engine that allows AI to modify software without ever creating invalid states. It enforces Propose → Validate → Commit → Timeline and enables safe AI-driven refactoring and multi-agent coding.
Open-source core: GitHub - jaityagi11/safechange-core
Would love feedback from anyone working on AI agents or safety.
Hey, @Jai_Tyagi, welcome to the community!
I moved your post to the Community category. We ask that you keep updates to your project in this thread, so it’s easier for everyone to stay up to date.
Any big hurdles you ran into that could help other developers? We’ve got a really great community here. Hope you enjoy your stay.
1 Like