Customer and agentic identity platform company Descope Inc. today unveiled Agentic Identity Hub 3.0, the latest version of its platform for securing artificial intelligence agents.
This time the focus is on the agents employees build or adopt inside their own organizations. The release goes after what Descope describes as a shadow AI problem. Staff are adopting and building agents faster than security and information technology teams can inventory them, and many of those agents run on shared, long-lived credentials with no recorded owner.
Descope cites Verizon Communications Inc.’s 2026 Data Breach Investigations Report for the scale of it. Regular AI use on corporate devices has spread to 45% of employees, the report found. Roughly two-thirds of them sign in to AI services with a personal account.
To get agents on the books, Descope now makes employees ask for them through an Agent Onboarding Portal. Each request starts in a self-service catalog and spells out what the agent will connect to, down to individual data sources and Model Context Protocol servers. Nothing is provisioned until a reviewer signs off. Every approved agent comes out of the process with an accountable owner and scoped access, and Descope said the workflow is quick enough that teams have no reason to route around it.
Many of the agents employees use were built by someone else, and products such as Anthropic PBC’s Claude, Anysphere Inc.’s Cursor and third-party MCP servers cannot be instrumented by the organizations running them. Security teams are left with no place to apply policy or capture a log. Version 3.0 integrates natively with MCP gateways, among them the Agentic AI Foundation’s open-source agentgateway project. That puts a single identity and policy layer in front of every agent and MCP server in use, including outside clients such as OpenAI Group PBC’s ChatGPT.
The update also targets static client secrets. Those secrets tend to leak into code repositories and rarely get rotated. Under the new client authentication options, an agent can sign a short-lived assertion with a private key that never leaves the client.
Internal agents get something close to single sign-on. The sign-in happens once, through whatever identity provider the company already uses, and access to each MCP server follows from that with no further login or consent screen. Descope said policy is evaluated on every request.
An administrator who wants to pull back an agent’s access can do it from one place, and no credentials have to be reissued. Behind the feature are identity assertion grants, the mechanism that also underpins Cross-App Access. Descope began supporting that open protocol in September.
Descope co-founder and Chief Executive Slavik Markovich said much of the hub’s early traction came from customer-facing agents and MCP servers. The same customers kept raising a different problem on the inside, he said, with employees “spinning up agents faster than anyone could track them.”
The Agentic Identity Hub handles authentication and access control for hundreds of MCP servers and more than 1 million agent identities, according to the company.
Descope is a venture capital-backed startup that closed a $35 million extension of its seed round in September 2025, taking its total funding to $88 million. Notable Capital, Lightspeed Venture Partners, Dell Technologies Capital and Unusual Ventures are among its investors.