Teleport Beams Trusted Runtime for AI Agents Reaches General Availability
SALT LAKE CITY -- Teleport announced during KubeCon + CloudNativeCon North America 2026 the general availability of Teleport Beams, a trusted runtime for securing AI agents in production infrastructure. Beams had been in public beta since March 2026 and connects agents to cloud accounts, Kubernetes clusters, PostgreSQL databases, MCP servers The post Teleport Beams Trusted Runtime for AI Agents…
Salt Lake City, KubeCon + CloudNativeCon North America 2026 -- Teleport unveiled the public release of Teleport Beams, a secure runtime for managing AI agents in production environments. Beams had been under evaluation by users since March 2026 in a beta phase. This new system enables AI agents to access cloud accounts, Kubernetes clusters, PostgreSQL databases, MCP servers, and other resources without exposing sensitive credentials.
Beams manages agent operations through four tiers of controls. Each runtime includes an identity granted by a user or other agent, and agents begin with no inherent permissions. They submit execution plans and access requests for review by a human or an artificial intelligence system before the separate runtime carries out the approved tasks.
The system automatically detects available resources and authenticates using standards like IAM, Kerberos, and X.509. An LLM gateway determines model selection and logs activity. To manage multiple agents, Beams identifies goal manipulation and coordinated behavior, scoring each session against the MITRE ATT&CK framework.
Each agent operates in short-lived virtual machines with controlled network connections. The system includes natural-language classifiers that evaluate agent activity against seven preset rules, such as detection of exposed secrets, destructive actions, and privilege escalation, along with support for custom rules. The Phases feature snapshots session artifacts and memory so work can resume or branch into parallel tasks. A kill switch can instantly end an agent's activity and all associated production connections.
Built on Teleport's Infrastructure Identity Platform, which is implemented at over 700 organizations, Teleport Beams addresses the challenge many enterprises face in securely connecting AI agents to production infrastructure. According to Teleport CEO Ev Kontsevoy, "Most enterprises have not connected agents to their production infrastructure.
They are blocked by a lack of delegated identity, strong access control without secrets and with full audit and traceability of intent to action." He added, "When the cost of an error is high, traditional sandboxes don't cut it." Teleport Beams is now available for widespread use.
Written by urgent.news from Container Journal's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.