name: CRI-O Status Capabilities description: >- Workflow capabilities exposed by the CRI-O HTTP status API for runtime introspection, container lifecycle control, and golang debugging. url: https://github.com/cri-o/cri-o version: '1.0' modified: '2026-04-28' api: CRI-O Status API baseURL: http://localhost capabilities: - name: Runtime Introspection description: >- Retrieve daemon configuration, storage driver, cgroup driver, and runtime defaults to understand how CRI-O is configured on a node. operations: - getInfo - getConfig inputs: - cri-o socket path outputs: - daemon info JSON - active TOML configuration - name: Container Inspection description: >- Look up runtime details for a specific container by ID, including its sandbox, image, PID, and creation time. operations: - getContainer inputs: - container ID outputs: - container info JSON - name: Container Pause Control description: >- Pause and unpause running containers without killing them, useful for debugging, snapshotting, or coordinating workloads. operations: - pauseContainer - unpauseContainer inputs: - container ID outputs: - HTTP 204 on success - name: Daemon Debugging description: >- Dump goroutine stacks for the running CRI-O daemon to diagnose deadlocks, hangs, and unexpected runtime behaviour. operations: - getGoroutines inputs: - cri-o socket path outputs: - goroutine stack trace useCases: - name: Node Troubleshooting description: >- Inspect storage and runtime configuration when diagnosing node-level Kubernetes container failures. capabilities: - Runtime Introspection - Container Inspection - name: Live Debugging description: >- Pause a misbehaving container and dump CRI-O goroutines while collecting forensic data, then unpause to restore service. capabilities: - Container Pause Control - Daemon Debugging - name: Operational Tooling Integration description: >- Build internal tooling that surfaces per-node CRI-O state for fleet operators outside of kubelet and crictl. capabilities: - Runtime Introspection - Container Inspection