mailsoftly-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill references an external MCP server at https://rube.app/mcp and documentation at composio.dev. These sources are not on the trusted list, though they are the legitimate endpoints for the Rube service.
  • [REMOTE_CODE_EXECUTION] (LOW): The use of RUBE_REMOTE_WORKBENCH and RUBE_MULTI_EXECUTE_TOOL involves executing tasks on remote infrastructure. This is consistent with the skill's primary purpose but represents a capability surface for executing arbitrary remote tool logic.
  • [Indirect Prompt Injection] (LOW): The skill processes external tool schemas and data without explicit boundary markers or sanitization. Evidence: 1. Ingestion points: RUBE_SEARCH_TOOLS results and tool outputs. 2. Boundary markers: Absent. 3. Capability inventory: Remote tool execution via Composio. 4. Sanitization: None mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:41 PM