payment-integration

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions designed to bypass safety filters or override agent constraints were found.
  • [Data Exposure & Exfiltration] (SAFE): No sensitive credentials, private keys, or file paths were identified. The skill explicitly instructs the agent never to log sensitive card data.
  • [Unverifiable Dependencies] (SAFE): The skill does not perform any external downloads or install unverifiable packages; it recommends using official SDKs.
  • [Indirect Prompt Injection] (SAFE): While the skill processes external data via webhooks, it includes comprehensive security checklists and instructions for PCI compliance to mitigate ingestion risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:33 PM