outreach-personalization-research

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow calls Amplemarket enrichment/search tools (mcp__claude_ai_Amplemarket__enrich_person, ...enrich_company, ...search_people) which can ingest and return outsider-authored web/social/profile text about third parties (e.g., LinkedIn/public bios, scraped pages) into the agent’s LLM context for synthesis.