agenty

Warn

Audited by Snyk on Mar 4, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly instructs using Membrane to run Agenty actions and proxy arbitrary Agenty API endpoints (see "Working with Agenty" and "Proxy requests"), which returns scraped/open-web job results and other third-party content that the agent is expected to read and could contain untrusted user-generated instructions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 4, 2026, 08:55 AM