The system category for agentic runtime security.
AARM defines the security controls an AI agent runtime must implement before any action is executed — intercept, evaluate against policy, decide, and produce a tamper-evident record.
Two conformance levels
Clear requirements for products serious about AI agent security.
All six requirements are MUST. Satisfying these is the baseline for AARM conformance — pre-execution interception through identity binding.
View requirements →Core plus three SHOULD requirements: semantic drift tracking, telemetry export, and least-privilege enforcement.
View requirements →Conformant builders
Products that satisfy AARM specification requirements.
A unified platform to secure and govern your AI and agents, delivering enterprise-grade protection.
One platform for MCPs, Skills, and Agents, with purpose-built security, fine-grained governance, and complete observability.
Formal enforces least-privilege at the wire protocol layer for humans and AI agents.
Discover, Detect, and Defend your AI, Agents, and MCP in real-time.
Enterprise governance platform for AI agents and MCP servers.
11 threat classes addressed
AARM systems are designed to defend against all known classes of attack on agentic AI.
Join the AARM Working Group
A system category specification built by security practitioners, researchers, and builders. Come shape the future of AI agent security.
Join the CSA Working Group