Noma AI Agent Security Platform

Israel flag

Noma AI Agent Security Platform is a governance-focused product that discovers, monitors, and defends enterprise AI systems and autonomous agents.

Security Posture Management
Are you the owner?Claim and verify your listing

License

Commercial

Category

GRC

Product description

Noma AI Agent Security Platform targets threats that conventional security tools may not detect, such as prompt injection, model poisoning, and agents that behave unexpectedly. It covers AI applications and autonomous agents across their lifecycle and addresses risks that spread between connected systems. The platform includes AI security posture management alongside automated threat protection.

Asset discovery covers models, agents, pipelines for data, and MLOps tooling, and it generates an AI and machine learning bill of materials for inventory. Continuous scanning and automated red teaming test these assets. Runtime guardrails detect and block malicious prompts and jailbreak attempts, along with agent operations that fall outside policy, and supply chain scanning checks for compromised models and MCP servers. Anomaly detection, agent behavior monitoring, and sensitive data leakage prevention are also included.

Compliance automation maps controls to OWASP's LLM top ten list, MITRE ATLAS, and the EU AI Act. The product is a commercial offering.