Software delivery platform company Akuity Inc. today introduced Agentic Control Plane, a layer that lets artificial intelligence agents read its pipeline data and act on it under the permissions the platform already enforces.
A second product, Akuity MCP Server, connects agents to the platform over the Model Context Protocol.
Teams are writing more code than they can safely ship, according to Akuity, and agents are already operating inside production environments whether or not anyone sanctioned them. Many of them connect through open-source MCP servers. Akuity said those servers hand over raw access, with no identity carried through and no audit trail of what the agent touched.
Akuity’s answer is to route every request an agent makes through Agentic Control Plane, which authenticates it as the specific person who connected the agent. No shared service account is involved. That person’s existing permissions travel with the connection across every instance and cluster Akuity manages, so there are no separate credentials to issue or rotate. A policy layer sits above all of it and can block sensitive production actions the user is otherwise cleared to perform.
What agents reach through the control plane is the operational record of the pipeline itself. Akuity said that covers deployment history, cluster health and change lineage across a customer’s fleet, down to the commit behind any given deployment.
Engineers can put one question to an agent and get an answer covering every cluster rather than checking each in turn. An agent aimed at an empty repository can stand up the environments and promotion rules for a new service in an afternoon, then promote the release from a chat window, with the same approval gates applying whether a person or an agent made the request. Akuity’s On-Call Agent takes the harder cases, pulling logs and recent changes to suggest a fix.
Major League Baseball is an early user. Michael Goodness, a principal DevOps engineer at the league, said the control plane surfaced more than 100 degraded applications within hours of being switched on, then traced a systemic issue in one project to its root cause in under 10 minutes.
“That’s exactly the workflow we wanted: Identify the problem, tell us if it’s systemic and hand us the fix,” Goodness said. “This is already bringing value to the team, and we are looking to expand its adoption.”
Akuity co-founder and Chief Executive Hong Wang said agents have moved past writing code and into “how infrastructure changes get delivered and promoted to production.” The winners of the next phase will not be whoever adopts AI fastest, he added, but whoever gives their agents a reliable path to do that work.
Akuity was founded by the creators of Argo CD and Kargo, the open-source projects for continuous delivery and environment promotion on Kubernetes. Lead Edge Capital led a $20 million Series A for the company in 2022. AI automations landed on the platform last September. On-call engineers are never paged for the Kubernetes application incidents the software catches and fixes on its own.
