media-use
Fail
Audited by Snyk on Nov 14, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (medium risk: 0.30). The skill directs the agent to pipe a remote install script from static.heygen.ai into bash and to call an undocumented, hidden
asset searchcommand, which is a concealment-style risk but shows no sign of intent to attack or defraud the user.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required
resolveworkflow sends the agent's intent to the HeyGen audio and asset catalogs, which return third-party-published items, and writes theirdescription/namefields into the manifest andindex.mdthat the agent reads; the agent cannot force delivery of any particular item, so this is low rather than medium.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (low risk: 0.30). The skill mainly resolves media files into a project directory, but it tells the agent to run a remote
curl | bashinstaller and self-update a CLI, which changes machine state without sudo and without verification, so I'm flagging it at low severity.
Issues (3)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata