salvo-data-extraction

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly parses and uses untrusted, user-generated HTTP request content (request bodies/JSON, form data, query parameters, headers, and path params) as part of its runtime extraction workflow (see JsonBody, parse_json/parse_form, query/param/header examples), so external third-party input can materially influence processing.