The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the kubectl command-line tool via Bash to manage Kubernetes deployments, including applying manifests and checking rollout status. These commands are executed locally within the cluster context provided by the user.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from a Kubernetes cluster by listing pods and reading deployment configurations, which represents an ingestion surface for potentially untrusted data.
Ingestion points: Data is retrieved from the cluster via k8sApi.listNamespacedPod and appsApi.readNamespacedDeployment in SKILL.md.
Boundary markers: None present; the skill treats retrieved Kubernetes metadata as trusted data.
Capability inventory: The skill has Write, Edit, and Bash(kubectl:*) capabilities, allowing it to modify the cluster state based on the ingested data.
Sanitization: No explicit sanitization or validation of pod labels or node selector values is performed before processing.

coreweave-upgrade-migration