The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute kubectl commands. These commands are limited to read-only diagnostic operations, such as checking rollout status, pod readiness, resource usage, and HPA configurations.
[DATA_EXFILTRATION]: While the skill retrieves logs and events that could contain sensitive information, it includes a proactive mitigation step. It uses sed to filter and redact sensitive patterns (e.g., passwords, tokens, credentials) from all command output before displaying it to the user.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it processes untrusted data from external sources (Kubernetes logs and events).
Ingestion points: Kubernetes container logs (kubectl logs) and cluster events (kubectl get events) as defined in SKILL.md.
Boundary markers: Findings are presented in a structured health dashboard table.
Capability inventory: The skill utilizes Bash (for kubectl execution), Read, Grep, and Glob tools.
Sanitization: Implements keyword-based redaction for secrets, although it does not specifically filter for malicious natural language instructions embedded in logs.

rollout-check