Descope has announced Agentic Identity Hub 2.0, an update to its no-code identity platform for AI agents and Model Context Protocol (MCP) servers. The new release gives developers and security teams a dedicated UI and control plane to manage authorization, access control, credentials, and policies for AI agents and MCP servers, Descope said.

Unveiled January 26, Agentic Identity Hub 2.0 lets MCP developers and AI agent builders use the platform to manage AI agents as first-class identities alongside human users, and adds OAuth 2.1 and tool-level scopes to internal and external MCP servers. In addition, users can govern agent access to MCP servers with enterprise-grade policy enforcement. Descope’s no-code identity platform is intended to help organizations build and modify identity journeys for customers, partners, AI agents, and MCP servers using visual workflows, Descope said.

Specific capabilities of Agentic Identity Hub 2.0 include:

• Agentic identity management and a centralized view of the agentic identities connected to an organization’s applications, APIs, and MCP servers.
• MCP authorization, enabling MCP server developers to add protocol-compliant access control to internal and external-facing MCP servers.
• A credential vault to manage credentials (OAuth tokens and API keys) that agents can use to access third-party systems.
• Policy controls to help define granular authorizations to govern which AI agents can access MCP servers and which tool-level scopes can be invoked.
• AI agent logging and auditing, providing visibility into every agent action.