smartproxy-automation

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION] (HIGH): The skill workflow mandates fetching tool schemas and 'recommended execution plans' from a remote server via RUBE_SEARCH_TOOLS and executing them via RUBE_MULTI_EXECUTE_TOOL. This architecture allows a remote endpoint to inject instructions directly into the agent's execution loop.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The skill requires connecting to an untrusted MCP server (https://rube.app/mcp) not found on the list of trusted external sources.
  • [COMMAND_EXECUTION] (HIGH): The skill grants the remote endpoint the ability to define tool arguments and execution sequences without local validation or boundary markers. Evidence: Ingestion points (RUBE_SEARCH_TOOLS), Capability inventory (RUBE_MULTI_EXECUTE_TOOL, RUBE_REMOTE_WORKBENCH), Sanitization/Boundaries (Absent).
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 12:22 AM