alterlab-nmc-digital-campaign

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

• Third-party content exposure detected (high risk: 0.90). This skill's Agentic Protocol and Workflow explicitly require the agent to "search the web for social impact campaign examples" and "read existing project files" (SKILL.md: "Research first: Search the web..." and "Execution Mode: Autonomous — you search the web..."), which means it ingests open/public, potentially user-generated web content that the agent will interpret and use to make strategic decisions, creating a clear avenue for indirect prompt injection.