mode-switcher

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

• Third-party content exposure detected (high risk: 0.80). The skill's mode definitions (notably research-mode and marketing-mode) explicitly direct the agent to use web search MCPs, perform SERP/backlink analysis, "cite every claim" and "search broadly," which requires fetching and interpreting untrusted public web content (modes/research.md and modes/marketing.md).