The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the superset CLI tool to perform diagnostic operations, such as checking service status, listing hosts, and verifying versions. It also provides remediation commands like starting the service or updating the CLI, which are consistent with standard vendor-supported maintenance tasks.
[DATA_EXFILTRATION]: The skill identifies application logs at ~/Library/Logs/Superset/main.log as a source for troubleshooting. To prevent unauthorized data exposure, the instructions explicitly require the agent to seek user permission before processing or sharing any log content, noting that these files may contain sensitive project names or file paths.
[PROMPT_INJECTION]: The ingestion of local application logs constitutes a surface for indirect prompt injection.
Ingestion points: The skill reads diagnostics from ~/Library/Logs/Superset/main.log.
Boundary markers: The agent is instructed to ask for explicit user consent before reading or reporting log content, which serves as a manual validation step.
Capability inventory: Local execution of superset CLI commands (status, auth, hosts, start, update).
Sanitization: No automatic sanitization of log data is implemented.

superset-doctor