procfu-automation

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly requires calling RUBE_SEARCH_TOOLS against an external Rube MCP endpoint (e.g., https://rube.app/mcp) to fetch third-party tool schemas and connection data which the agent must read and follow to select and execute tools, so untrusted remote content can materially influence actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill requires adding and calling the runtime MCP endpoint https://rube.app/mcp, which the agent queries (RUBE_SEARCH_TOOLS/RUBE_MULTI_EXECUTE_TOOL) to fetch tool schemas and to invoke remote tools—meaning remote content controls the agent's instructions and triggers remote execution.