Skip to main content

AgentGuard

AgentGuard is a real-time guardrail for agent actions during runtime. It flags and blocks malicious or policy-violating tool calls based on configurable policy sets — supporting standard frameworks (EU AI Act, GDPR, OWASP LLM Top 10) as well as custom policies.

Key Capabilities​

  • Runtime enforcement — evaluate every tool call against a policy set before it executes
  • Standard policy support — EU AI Act, GDPR, OWASP LLM Top 10, and more out of the box
  • Custom policies — upload your own rules as PDF or TXT
  • Monitor & audit — dashboard with violation statistics, risk distributions, and per-action activity logs
  • Flexible integration — connect via the Virtue Gateway for full protection, or use the AgentGuard SDK standalone in any agent framework