apollo-to-cargo

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime path uses first-party, user-supplied Apollo CSV data and then calls provider connectors (apolloio, waterfall) via Cargo, so the LLM ingests outsider-authored free text originating from the user’s export/supplied rows (e.g., names/domains/emails) without selecting a trusted specific item first.