Agent Audit

I built AgentAudit after realizing something scary: as we rely more on AI agents to send emails, schedule meetings, and automate work, we often have no visibility into what they’re actually doing. AgentAudit is a monitoring and audit layer for AI agents. Today it connects to Gmail and Google Calendar to flag risky behavior like: ⦁ emails sent to new or unknown domains ⦁ actions outside business hours ⦁ unusual recipient patterns The goal is simple: catch problems before your agents create real damage. This is just the beginning. I’m actively working on expanding monitoring to more tools, agents, and workflows. I’d love your feedback, ideas, and criticism.