The Agent Skills Directory

[SAFE]: The skill provides legitimate guidance and commands for Kubernetes cluster administrators and SREs to troubleshoot common issues.- [COMMAND_EXECUTION]: The skill utilizes the kubectl CLI to perform introspection and management of cluster resources. This includes potentially destructive commands such as kubectl delete pod ... --force and kubectl scale, which are provided in the context of resolving stuck resources.- [PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection by interpolating user-provided inputs into shell commands.
Ingestion points: User-supplied variables such as <pod-name>, <namespace>, <service-name>, and <container-name> are used across various commands in SKILL.md.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present around the variable placeholders.
Capability inventory: The skill uses kubectl with the ability to read logs, describe resources, run pods (kubectl run), and modify deployments (rollout, scale, delete).
Sanitization: No sanitization or validation of the user-provided strings is performed within the skill instructions.- [CREDENTIALS_UNSAFE]: Provides a template for kubectl create secret docker-registry. The command correctly uses placeholders like <user> and <password> instead of hardcoded credentials, aligning with best practices for secret management.

k8s-debug