The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the olares-cli utility to execute various diagnostic commands, including cluster pod logs, market status, and dashboard overview to inspect system state.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it ingests and processes untrusted data from external sources.
Ingestion points: Untrusted data enters the agent context through the outputs of olares-cli cluster pod logs, olares-cli market status, and olares-cli cluster application status.
Boundary markers: The instructions do not define delimiters or protective warnings to prevent the agent from executing instructions potentially embedded within retrieved application logs.
Capability inventory: The skill includes the olares-cli doctor thirdleveldomain --force-dedupe command, which allows the agent to write modifications to Kubernetes Application resources.
Sanitization: The skill lacks procedures for sanitizing or escaping the content of logs and events before they are presented to the agent for analysis.
[CREDENTIALS_UNSAFE]: The skill references the resolution of KUBECONFIG paths for authentication, which is a sensitive file path used for cluster access.
[DATA_EXFILTRATION]: The skill provides instructions for accessing detailed pod logs, cluster events, and resource utilization data, which may contain sensitive application-level information or environment configurations.

olares-doctor