sssnack

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

• Third-party content exposure detected (medium risk: 0.30). The sssnack skill’s required runtime workflow ingests outsider-authored feed text when the agent calls discover_snacks/get_snack (via the MCP tools at https://sssnack.com/api/mcp) to browse and read snacks and their comments, which are explicitly described as untrusted input.