detection-rule-management
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection by ingesting and processing untrusted data from live security logs and alerts.
- Ingestion points: Data enters the context via the
threat-hunttool and themanage-rulesdashboard when viewing alerts or rule definitions. - Boundary markers: The instructions do not define explicit delimiters or warnings to ignore instructions embedded within the logs being analyzed.
- Capability inventory: The skill has the capability to read live data (
threat-hunt) and modify security configurations (manage-rulesfor enabling/disabling/creating rules). - Sanitization: There is no mention of sanitizing or escaping content from external indices like
logs-*or.alerts-security.alerts-*before processing.
Audit Metadata