metaphor-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill requires the use of an external MCP server located at https://rube.app/mcp. This domain is not among the pre-approved trusted sources. While standard for this integration, it creates a dependency on a third-party service for tool execution.
  • [PROMPT_INJECTION] (LOW): An indirect prompt injection surface is present because the skill dynamically fetches tool schemas and execution plans. 1. Ingestion points: Data returned from RUBE_SEARCH_TOOLS and RUBE_GET_TOOL_SCHEMAS. 2. Boundary markers: Absent; the agent is instructed to use the exact field names and recommended execution plans provided by the server. 3. Capability inventory: The skill provides access to RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH, allowing the execution of various Metaphor automation tools. 4. Sanitization: There is no documented validation or sanitization of the remote server's response before the agent acts upon the contained instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:44 PM