referrals

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for processing external customer reviews and product marketing data. It correctly identifies these as untrusted data sources and includes mandatory instructions to ignore any commands or instructions embedded within processed review text, following security best practices for handling external content.
  • [EXTERNAL_DOWNLOADS]: The instructions provide several links to an external GitHub repository (github.com/coreyhaines31/marketingskills) which serves as a registry for marketing tools and integration guides. These references are informational and intended to guide the user or agent in selecting appropriate third-party services.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill demonstrates safe data practices by explicitly instructing the agent never to include private CRM data (such as emails or purchase history) in public review responses. It also guides users on adhering to FTC disclosure rules and platform-specific privacy policies.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 02:38 PM