skyline

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions require the agent to ingest untrusted data from external sources such as email and SMS to retrieve authentication codes and links. This creates a surface for indirect prompt injection attacks where a malicious message could influence agent behavior.
  • Ingestion points: Accesses short-lived MFA codes and confirmation links from Gmail, Messages/SMS, and browser content.
  • Boundary markers: The instructions do not specify the use of delimiters or clear warnings to ignore potential instructions embedded within the retrieved MFA communications.
  • Capability inventory: The skill allows the agent to perform account searches, authentication, account creation, and password management through the Skyline service.
  • Sanitization: There are no requirements for sanitizing or validating the contents of the external communications before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 01:52 AM
Security Audit — agent-trust-hub — skyline