company-contact-finder

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md Steps 2 and 5) calls external people-search tools—mcp__gooseworks__crustdata_nl_search, mcp__gooseworks__crustdata_search, and mcp__gooseworks__sixtyfour_nl_search—to ingest profiles and LinkedIn URLs from external/public people databases, which the agent reads and uses to decide follow-up actions and outputs, exposing it to untrusted third-party content that could carry indirect prompt-injection.