The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted Kubernetes log data, which may contain malicious instructions intended to influence agent behavior.
Ingestion points: The tools search-logs and analyze-logs return raw and aggregated log data (specifically the body field) to the agent context.
Boundary markers: The skill documentation does not provide delimiters or specific instructions for the agent to distinguish between log content and its own instructional set.
Capability inventory: The tools provided in this skill are limited to read-only log retrieval and analysis, lacking high-privilege operations like file writing or remote code execution.
Sanitization: No sanitization, escaping, or validation logic is defined for the log data before it is ingested and processed by the agent.

kubesense-logs