detecting-shadow-it-cloud-usage

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection. \n
  • Ingestion points: Ingests untrusted destination domains and traffic data from proxy access logs and DNS query logs (SKILL.md). \n
  • Boundary markers: Absent. The skill lacks instructions to wrap log data in delimiters or warnings to ignore embedded instructions. \n
  • Capability inventory: The skill uses Python pandas for traffic pattern analysis, domain classification, and risk scoring. \n
  • Sanitization: Absent. No procedures are defined for sanitizing or validating the contents of the logs before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 11:45 AM
Security Audit — agent-trust-hub — detecting-shadow-it-cloud-usage