Agent Identity & Governance
Agents are showing up in enterprises with access to data and APIs nobody expected them to have. In this conversation, Solo.io Global Field CTO Christian Posta explains why agents don’t fit human or workload identity models. He covers why basic zero trust practices are now essential, and how approaches like Microsoft Entra, AgentCore, SPIFFE and AAuth answer the questions enterprises need answered: who an agent is, what it has done, what it’s about to do, and who’s accountable. He also explains why approval for consequential actions has to be enforced outside the agent, with policy that flags sensitive actions in flight.
Try it yourself
Lock down which agents can call which tools, and see every action they take, in free hands-on labs.
- Free lab: agentgateway: Credentials and Access Control
- Free lab: Secure Your MCP Servers with OAuth
- Free lab: Multiplex MCP Servers & Control Auth Policy
- On-demand workshop: MCP Security in the Enterprise
Putting agents into production?
Talk to Solo.io about agent identity, MCP governance and human-in-the-loop controls for your environment. Book a meeting →
