research

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

• Third-party content exposure detected (high risk: 0.90). The skill launches external "researcher" agents (perplexity-researcher, claude-researcher, gemini-researcher) to collect and synthesize results with source attribution for research/current-events, which clearly entails fetching and reading open/public third-party web content (untrusted public sites) as part of its workflow.