social-graph-ranker

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

• Third-party content exposure detected (high risk: 0.80). The workflow explicitly requires pulling the user's graph "from X, LinkedIn, or both" (SKILL.md step 2), which ingests user-generated social data from public third-party platforms that the agent reads and uses to rank and drive outreach decisions, creating a clear vector for indirect prompt injection.