designing

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

• Third-party content exposure detected (high risk: 0.80). The skill's research hooks (notably "research-verify" in the Research Hooks table) explicitly include a "Web" agent and Phase 1/Phase 3 execution blocks spawn Task() agents for research, which indicates the agent will fetch and consume public web content (untrusted, user-generated) as part of its workflow.