bigpicture-io-automation

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill utilizes a dynamic discovery mechanism via RUBE_SEARCH_TOOLS that fetches execution plans and tool schemas from a remote service at https://rube.app/mcp. This creates a surface for indirect prompt injection where untrusted data from the external service could potentially manipulate the agent's logic or execution steps.
  • Ingestion points: Results from RUBE_SEARCH_TOOLS defined in SKILL.md.
  • Boundary markers: Not present; instructions direct the agent to follow returned schemas and execution plans directly.
  • Capability inventory: Includes tool execution via RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH for Bigpicture IO operations.
  • Sanitization: No sanitization or validation of the remote schemas is documented in the skill.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute operations using RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH based on schemas retrieved at runtime from an external source.
  • [EXTERNAL_DOWNLOADS]: The skill references an external MCP server endpoint (https://rube.app/mcp) which is required for its core functionality and acts as a remote repository for tool configurations.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 02:38 AM