Secure

AI agents interact with enterprise systems, tools, data, and other resources to perform tasks. As these interactions expand, organizations need controls over what agents can access, what actions they can perform, and how those interactions are secured.

Agent Identity Security provides runtime controls to govern these interactions. Organizations can define and enforce agent access through Policies and secure agent-to-MCP communication through the MCP Gateway.

Together, these capabilities help organizations apply appropriate security controls as agents operate across enterprise environments.

Policies

Policies define runtime controls for agent access and actions across supported resources.

Organizations can use policies to control agent interactions with resources such as MCP tools, files, terminal commands, web resources, packages, and version control systems. Depending on the supported resource and action, policies can allow, deny, or require approval for an agent action.

Learn more: Policies

MCP Gateway

MCP Gateway provides a governed path for communication between AI agents and MCP servers.

It enables organizations to apply authentication, access policies, data protection, and visibility to supported agent-to-MCP interactions.

Learn more: MCP Gateway