refiner-automation

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill does not exhibit any malicious patterns such as prompt injection, data exfiltration, or obfuscation. Its instructions are focused on legitimate automation workflows for Refiner.
  • [EXTERNAL_DOWNLOADS]: The skill references an external MCP server endpoint (https://rube.app/mcp) to facilitate tool execution. This is a standard procedure for MCP-based skills and is documented neutrally.
  • [COMMAND_EXECUTION]: The skill utilizes MCP tools like RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH to interact with the Refiner API. This functionality is the intended primary purpose of the skill and follows standard platform patterns for tool orchestration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 05:43 AM