crowdin-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • External Dependency (LOW): The skill requires the user to add an external MCP server at https://rube.app/mcp. While this is the intended mechanism for the skill's automation, the source is not on the trusted provider list, making it an unverifiable external dependency.
  • Indirect Prompt Injection Surface (LOW): The workflow relies on RUBE_SEARCH_TOOLS to fetch execution plans and tool schemas at runtime.
  • Ingestion points: Data is ingested from the RUBE_SEARCH_TOOLS response, including recommended execution steps.
  • Boundary markers: Absent. The skill instructions encourage the agent to follow returned schemas and plans directly.
  • Capability inventory: Uses RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH which can perform impactful operations on the Crowdin platform.
  • Sanitization: No validation or sanitization is mentioned for the instructions or schemas returned by the remote server.
  • Credential Handling (SAFE): The skill appropriately uses RUBE_MANAGE_CONNECTIONS for authentication rather than requesting hardcoded API keys or secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM