The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local shell script {baseDir}/diagnose.sh to run comprehensive system health checks. This is the primary mechanism for the diagnostic functionality and is consistent with the skill's description.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through the ingestion of external system data.
Ingestion points: The config and environment categories involve reading local configuration files and environment variables.
Boundary markers: There are no delimiters or protective instructions provided to prevent the agent from interpreting data within those files as new instructions.
Capability inventory: The skill can execute shell scripts and access system-wide metadata and network status.
Sanitization: No sanitization or validation of the external configuration data is mentioned or documented.

diagnostic-runner