auth-specialist

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • SAFE (SAFE): No malicious patterns or behaviors were detected. The skill is properly focused on security auditing for authentication flows.
  • Indirect Prompt Injection (LOW): The skill processes untrusted user-provided code for review, which creates a potential surface for indirect prompt injection. Ingestion points: User-submitted code snippets for authentication review. Boundary markers: None provided in the prompt instructions to isolate user content. Capability inventory: Limited to providing security advice and fix actions based on defined regex patterns. Sanitization: No evidence of input sanitization logic for the reviewed code.
  • Metadata (SAFE): The metadata correctly identifies the skill as an authentication specialist and uses descriptive tags for appropriate discovery.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:11 PM