Codenotary Launches AgentMon 3 with Adaptive Runtime Security Policies; Expands Availability on AWS Marketplace

Enterprise AI security platform dynamically adapts to customer behavior, protects more than 5 million AI agent interactions every day

HOUSTON--(BUSINESS WIRE)--Codenotary, leaders in assuring safe and secure use of AI, today announced AgentMon 3, the latest generation of its enterprise AI security platform, introducing adaptive runtime security policies. These continuously evolve as AI agents operate across an organization by learning from customer-specific workflows, observed behavioral patterns, and newly emerging threats.

The company also announced that AgentMon is now available through AWS Marketplace, making enterprise deployment significantly easier for organizations already operating on Amazon Web Services.

AgentMon has matured into one of the largest enterprise AI runtime security platforms in production today. The platform currently observes, analyzes, and secures more than 5 million AI agent interactions every day across enterprise customer environments. This operational scale has provided Codenotary with extensive real-world insight into how autonomous AI agents behave in production, enabling the company to build adaptive security capabilities based on actual enterprise deployments rather than laboratory simulations.

“As organizations rapidly deploy coding assistants, autonomous software engineering agents, business automation platforms, AI-powered customer support systems, and custom orchestration frameworks, traditional security models that are based on static allow-lists and manually maintained policies are proving increasingly inadequate,” said Moshe Bar, CEO and co-founder, Codenotary. “AgentMon 3 continuously learns from millions of real-world agent interactions while adapting to each customer's environment and incorporating intelligence from newly emerging threats. That enables organizations to secure AI at enterprise scale without creating an unsustainable operational burden."

AgentMon 3 redefines AI runtime security with an adaptive behavioral model built for autonomous AI agents. Every action an AI agent performs continuously shapes a live behavioral baseline, allowing organizations to distinguish normal operations from risky or anomalous behavior in real time.

Unlike conventional software, AI agents evolve constantly through new prompts, model upgrades, tool integrations, memory expansion, and workflow changes. Static security rules cannot keep pace. AgentMon replaces manual policy writing with dynamically generated, self-refining security policies based on how each organization actually uses AI. These living policies adapt automatically across teams, roles, agents, and workflows, creating precise security baselines unique to each environment.

By continuously learning from legitimate behavior, software updates, operational changes, and emerging threats, AgentMon eliminates much of the manual rule-tuning required by traditional security tools and cuts policy maintenance by up to 80%.

AgentMon 3 closes a critical security gap in autonomous AI systems. Many AI tools depend on built-in permission prompts and allow-lists that developers often weaken or disable for speed. Because AgentMon monitors actual runtime behavior independently of those controls, it continues to detect high-risk actions even when native safeguards are bypassed, misconfigured, or turned off.

Unlike signature-based security products, AgentMon continuously integrates organizational history, evolving agent capabilities, changing workflows, and emerging AI threat intelligence into its policy engine. As enterprise AI deployments mature, AgentMon’s adaptive security policies mature with them.

Every security decision is evaluated with deep context, including agent identity, permissions, historical patterns, data sensitivity, requested resources, prior human approvals, and live threat intelligence. This context-aware approach reduces false positives while improving detection of sophisticated AI-driven attacks.

AgentMon also correlates intent with impact. Detection is based on observed file access, network activity, credential use, process execution, and system connections – not on the agent’s own self-report. This makes AgentMon resilient to prompt obfuscation, multilingual attacks, and evasion techniques that bypass text-only security filters.

All runtime decisions made by AgentMon 3 are cryptographically recorded in Codenotary’s immutable tamper-proof ledger, creating a verifiable audit trail for compliance, investigations, and forensic analysis.

AgentMon 3 supports enterprise AI environments spanning coding assistants, autonomous software engineering agents, internal AI systems, orchestration frameworks, cloud-native services, and distributed multi-agent architectures. Available immediately worldwide and through Amazon Web Services Marketplace, AgentMon 3 gives enterprises a scalable, adaptive foundation for securing AI agents in production.

About Codenotary

Used by hundreds of customers worldwide – including the world’s leading banks, governments, and defense organizations – Codenotary delivers technology that enables secure and trusted agentic networks in the modern organization. For more information, visit www.codenotary.com.