investor-call-prep

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

• Third-party content exposure detected (high risk: 0.85). At runtime the workflow ingests outsider-authored free text from public web content via the “scrapegraph” step (scraping https://{firm_website}, /portfolio, /team) and from Google Calendar event fields (titles/descriptions/attendee emails) that are authored by third parties (event organizers/investors), which the agent then places into LLM prompts for thesis/fit generation.