Agents as platform users
AI agents are first-class platform users, able to act on or assist with developer workflows and platform engineering operations, not just source code.
Agents shouldn't have to reason in raw Kubernetes primitives or stitch together disconnected tools. OpenChoreo exposes the same abstractions, golden paths, and guardrails to AI agents that it gives to developers so agents can deliver, operate, and optimize software as first-class platform users.
AI agents are first-class platform users, able to act on or assist with developer workflows and platform engineering operations, not just source code.
Agents operate under the same access-controlled guardrails as humans, not separate unpredictable paths.
Built-in SRE and FinOps agents run inside the platform, and serve as the blueprint for building your own.
OpenChoreo exposes MCP servers for both the Control Plane and the Observability Plane, giving agents access to live platform context; components, environments, deployment state, logs, metrics, traces instead of just source code, disconnected tools and YAML.
Use the AI tools and agents you already use with OpenChoreo.
A growing ecosystem of reusable agentic skills packages common platform and developer tasks, ready to drop into your AI workflows.
OpenChoreo does not force AI to reason with Kubernetes primitives and disconnected tools. It exposes higher-level platform constructs for delivery and operations that agents can understand and act on efficiently.
This unified context, combined with strong RBAC, enables AI agents to safely perform and assist with real delivery and operational tasks at scale.