expanso-content-routing

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

• Third-party content exposure detected (high risk: 0.75). The skill routes inbound HTTP events from an http_server at runtime, so outsider-authored request body/text from external clients would be ingested by the pipeline mappings and thus could reach the agent’s LLM context if those mappings are passed into an LLM.