waterfall-enrich-contacts

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow calls an OUTSIDER enrichment provider adapter (e.g., scripts/providers/fullenrich.py, apollo.py, hunter.py, dropcontact.py) over the network and ingests the provider’s JSON responses (free-form text fields like names/titles/emails/phones) into the agent’s LLM context via the adapter output that is later used for audit/write decisions.