The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and analyze external, potentially untrusted data to diagnose issues.
Ingestion points: The agent reads pipeline schedules, asset dependencies, terminal output, and files within the local logs/ directory (specifically logs/runs, query logs, and export logs).
Boundary markers: There are no specific delimiters or instructions provided to the agent to treat external log content as data only or to ignore embedded instructions.
Capability inventory: The agent has the capability to execute shell commands via the bruin CLI, query databases, and run Python scripts for asset validation.
Sanitization: No sanitization, filtering, or validation steps are defined for the logs or pipeline metadata before they are processed by the agent.
[COMMAND_EXECUTION]: The skill makes extensive use of the bruin and bruin cloud CLI tools for diagnostic purposes (e.g., bruin cloud runs diagnose, bruin cloud instances failed-logs). These are recognized vendor-specific tools from 'bruin-data' and are appropriate for the skill's primary purpose.

freshness-check