coolhunter

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflow (see "Step 2: Web Research Execution" in references/workflow-execution.md, the SKILL.md Research Workflow, and workflow-plan-coolhunter.md) explicitly instructs the agent to perform web_search/web-browsing of recent public news URLs and to extract quotes, signals, and citations from those external sites for analysis and fact-checking, meaning untrusted third-party content is fetched and directly used to drive methodological choices and report actions.