israeli-startup-investment-analyzer

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow ingests outsider-authored free text provided by the investor/user (“a deck or data-room summary”); the model then uses those user-provided numbers/claims and templates to generate the memo, so an outsider can inject poison via the submitted pitch/deck text without the agent first selecting a specific item.