I have built Recursant, an AI agent governance platform that governs agents across stacks and clouds to maintain compliance. In many large enterprises this is a real compliance issue with different teams using different frameworks and runtime environments, which creates control risk.
I hope to get some feedback on it. It is a viable alternative to the large vendor controlled stacks. Architecturally it is quite unique given its use of the Istio / sidecar pattern. This enables agent isolation at the network layer.
Thanks Anders