iii-http-middleware

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

• Third-party content exposure detected (high risk: 0.70). The skill's middleware explicitly ingests and interprets arbitrary incoming HTTP request data (e.g., req.request.headers, path_params, query_params as shown in SKILL.md and the Reference Implementation) and uses it to decide actions (respond vs continue), which means untrusted third-party request content can materially influence control flow and thus could enable indirect prompt injection.