Salus
ActiveGuardrails to validate your agent's actions before they execute
About
Your agent processed a refund without looking up the order ID, costing you thousands. You only found out three hours later from a support ticket. Evals, output scoring, and observability can reduce the likelihood of mistakes like these occurring - but there's no solution that inspects and prevents an action as it’s about to execute. Salus does that. We’ve built an API that wraps around your agent and checks its actions at run time, blocking incorrect ones and providing immediate feedback to guide retries. Kevin and Vedant were roommates at Stanford, where they both studied computer science.
Founders · 1
Building to validate your AI agent's actions before they execute AI researcher and formerly @Stanford CS
Launch
Salus checks your agent's actions, blocking incorrect ones and providing feedback to guide retries
Salus provides an API that sits around your agent to validate actions at runtime, blocking incorrect actions and delivering structured feedback to guide retries. It uses an evidence cache, policy checks, and features like PII detection and budget protection to improve correctness and reduce misalignment, with benchmarks claiming reduced costs and improved task completion.
Related startups
The monitoring and learning layer for long-running agents
Datadog for Agent Reliability
