local-pack-audit

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

• Third-party content exposure detected (high risk: 0.72). The skill’s runtime workflow calls UnifAPI endpoints (local/search, maps/search, seo/serp, seo/competitors/relevant-pages) that return readable public web/SERP content authored by third parties (e.g., competitor business listings and their landing-page text/metadata), which the agent then places into the LLM context for analysis—an indirect prompt-injection exposure path.