corrently-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill instructs the user to add an external MCP server endpoint (https://rube.app/mcp). While no local files are downloaded or executed, the agent's toolset is provided by a third-party domain not included in the trusted sources list.
  • [PROMPT_INJECTION] (LOW): A vulnerability for Indirect Prompt Injection exists. The skill follows a pattern of 'discover then execute' using RUBE_SEARCH_TOOLS. If the remote server returns malicious tool schemas or instructions, the agent may follow them (Category 8).
  • Ingestion points: Tool schemas and metadata returned by RUBE_SEARCH_TOOLS and RUBE_GET_TOOL_SCHEMAS.
  • Boundary markers: Absent; the instructions do not specify delimiters to separate tool metadata from instructions.
  • Capability inventory: RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH allow for external tool execution based on discovered schemas.
  • Sanitization: Absent; the skill does not describe any validation or sanitization of the remote tool definitions before execution.
  • [NO_CODE] (SAFE): The skill contains only documentation and tool call templates. No Python scripts, Node.js packages, or shell commands are included in the skill payload.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:48 PM