hypercognitive-skill-compiler

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md Research Capabilities and mode_research_protocol explicitly authorize using searxng_searxng_web_search and searxng_web_url_read to fetch and validate information from the public web, meaning the agent will ingest and act on untrusted third-party content (public webpages) as part of its workflow.