mapbox-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No attempts to override system instructions or bypass safety filters were found.
  • [Data Exposure] (SAFE): No hardcoded credentials or sensitive file paths detected. The skill uses a managed connection tool for authentication.
  • [Remote Code Execution] (LOW): The skill facilitates tool execution via RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH. This is the intended functionality of the toolkit for automating Mapbox tasks through the Rube MCP server.
  • [Indirect Prompt Injection] (LOW): The skill possesses a vulnerability surface as it dynamically retrieves tool schemas and execution instructions from an external service.
  • Ingestion points: Tool schemas and execution plans are fetched at runtime via RUBE_SEARCH_TOOLS from rube.app.
  • Boundary markers: The instructions do not specify boundary markers for the data returned by the search tool.
  • Capability inventory: The skill can execute various Mapbox operations and remote workbench commands based on the external tool definitions.
  • Sanitization: The skill relies on the MCP client and the Rube service for validation of tool arguments.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:15 PM