The Agent Skills Directory

[COMMAND_EXECUTION]: The skill triggers the execution of local shell scripts for system diagnostics and log collection, specifically {baseDir}/scripts/diagnose.sh and deployment/collect-logs.sh.
[INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by ingesting and processing untrusted data from the live Kubernetes environment while maintaining command execution capabilities.
Ingestion points: The agent is instructed to read and analyze outputs from kubectl logs, kubectl get events, and file contents retrieved via kubectl exec (e.g., /tmp/health_check_alive).
Boundary markers: None. There are no instructions to use delimiters or ignore embedded natural language instructions within the retrieved log or event data.
Capability inventory: The skill has the capability to execute arbitrary commands within containers via kubectl exec and run local bash scripts on the host.
Sanitization: None. The skill does not implement sanitization or filtering of the cluster output before it is processed by the agent, though it does advise the agent to manually redact credentials if found.

debug-buttercup