summarise-granola

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow ingests Granola meeting transcript text fetched via the Granola API (scripts/granola.py get → build_transcript() → transcript markdown saved under data/transcripts/ → passed as “full raw transcript text” into the summary agent prompt), and that transcript is outsider-authored content (other meeting participants’ speech) that can contain indirect prompt-injection strings.