geoapify-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill requires the user to add an external MCP server endpoint (https://rube.app/mcp). This domain is not on the trusted organizations list. While necessary for the skill's primary function, it represents a dependency on an unverified third-party service.\n- [PROMPT_INJECTION] (LOW): The skill possesses an Indirect Prompt Injection surface (Category 8) by design.\n
  • Ingestion points: Untrusted data is ingested via RUBE_SEARCH_TOOLS, which returns tool slugs, input schemas, and execution plans.\n
  • Boundary markers: Absent; the agent is instructed to use the returned schemas directly without validation or delimiters.\n
  • Capability inventory: The skill has the capability to execute state-changing operations via RUBE_MULTI_EXECUTE_TOOL and bulk operations through RUBE_REMOTE_WORKBENCH.\n
  • Sanitization: No evidence of sanitization or schema verification is present in the skill instructions.\n- [REMOTE_CODE_EXECUTION] (LOW): The skill uses RUBE_REMOTE_WORKBENCH and run_composio_tool(), which are wrappers for executing remote capabilities defined by the external MCP provider. This allows the remote service to influence the agent's executable actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:48 PM