transcription-to-content

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly asks users to "paste transcript or key sections" and instructs the agent to read, mine, and transform those transcripts (e.g., full podcast transcripts), which are arbitrary user-provided or publicly sourced content and therefore untrusted third-party input the agent will interpret.