
Guardrails to validate your agent's actions before they execute
screenshot pendingSalus provides a runtime solution designed to enhance the safety and effectiveness of AI agents that interact with various backend tools. Positioned as a policy-aware proxy, Salus acts as an intermediary between AI agents and the tools they utilize, ensuring that each action taken by the agent is evaluated against established policies and evidence before execution. This approach helps mitigate risks associated with AI-driven actions, allowing organizations to maintain control over the outcomes of these interactions. Salus is compatible with a wide range of AI agents, including those from OpenA…