The Agent Skills Directory

[SAFE]: The skill is a legitimate instruction set for managing Kubernetes infrastructure using the vendor's MCP server. No malicious patterns, obfuscation, or unauthorized access attempts were found.
[EXTERNAL_DOWNLOADS]: The skill metadata references the official vendor repository on GitHub (kubesense-ai), which is a trusted source.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external Kubernetes environments (pod statuses, events, workload metrics). While this represents a data ingestion surface, the instructions emphasize using structured discovery tools and specific investigation patterns (Discovery Order), which helps manage the flow of external content into the agent's context. The skill instructs the agent to rely on specific names retrieved from previous tool calls, minimizing the risk of processing unvalidated inputs.

kubesense-infra