Building Relay a runtime security layer for AI agents because my own agent exposed live database credentials with zero warning. No checkpoint, no policy, nothing. Just decided and acted.
Curious if others have had close calls. What's the most dangerous thing an agent tried to do in your setup whether it got stopped or not?
Most agent security tools filter prompts before they reach the LLM.
Relay checks the action instead — sitting between your agent and its
tools, validating every GitHub call, shell command, database query, or deploy before execution.
Blocked calls return a clear reason.
Every decision is logged. Policies update from the dashboard without touching your agent code.
Open source, model-agnostic, with optional Docker sandboxing for shell agents.
