meeting-insights-analyzer

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

• Third-party content exposure detected (high risk: 0.75). The required workflow scans a user-provided folder for transcript files and then quotes “actual quote from transcript” into the LLM output context, so any outsider-authored meeting text included in those files (e.g., other participants’ speech) would be ingested as free-form text at runtime.