video1.0<iframe src="https://www.loom.com/embed/b790d4f7728b43afa7ae342a6c596176" frameborder="0" width="1110" height="832" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>8321110Loomhttps://www.loom.com8321110https://cdn.loom.com/sessions/thumbnails/b790d4f7728b43afa7ae342a6c596176-7de25bb1f4bd54b0.gif221.345In this video, I discuss how our AI agent can safely execute infrastructure changes to production while addressing security concerns. Axite prevents unauthorized changes by enforcing approvals and providing a clear evidence chain of what was executed. I demonstrate how the approval process works, ensuring that any discrepancies in the agent's retry will halt execution. This is crucial for maintaining accountability and security in our deployment process. Please review the approval workflow and ensure you understand your role in this process.