assemblyai

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

• Third-party content exposure detected (low risk: 0.10). The skill integrates with AssemblyAI via Membrane to manage transcripts and speech-to-text data, which may ingest user-generated content or transcript texts, but lacks any continuous background feed or monitoring workflow that automatically ingests outsider submissions without user selection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to install and run the Membrane CLI from an unofficial repository (GitHub membranedev/application-skills and npm @membranehq/cli), creating an unverifiable external operational dependency.