agent-reputation-benchmark

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

• Third-party content exposure detected (high risk: 0.95). At runtime the required workflow calls UnifAPI operations (local/search, maps/search, seo/serp) to fetch and sample public review text and local-pack/listing fields from third parties (outsider-authored Google/Maps reviews), which the agent then ingests into the LLM context for scoring (neighborhood-language %).