The Agent Skills Directory

[PROMPT_INJECTION]: The skill fetches and processes alerts from Monte Carlo Data, which constitutes an untrusted ingestion point for external data. This creates a surface for indirect prompt injection where malicious content within an alert could attempt to influence the agent's behavior during the triage process.
Ingestion points: Alert data is retrieved via the get_alerts tool as described in SKILL.md.
Boundary markers: The instructions do not explicitly define delimiters or specific warnings to isolate alert content from agent instructions.
Capability inventory: The skill utilizes tools with write capabilities, including update_alert, set_alert_owner, and create_or_update_alert_comment (SKILL.md).
Sanitization: No explicit sanitization or validation of the alert payload is mentioned in the prompt instructions.

automated-triage