sf-agent-api-variables

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow reads user-supplied free text because request “variables” (e.g., variables[].value with type: "Text") are provided in start-session/send-message bodies and can be included in the model prompt via includeInPrompt.