googlesheets-automation

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the configuration of an external MCP server at https://rube.app/mcp. This connects the agent to a remote tool provider, which defines and potentially executes the logic for the Google Sheets interactions.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Google Sheets using tools like GOOGLESHEETS_BATCH_GET and GOOGLESHEETS_LOOKUP_SPREADSHEET_ROW.
  • Ingestion points: Data is pulled from spreadsheet cells into the agent's context.
  • Boundary markers: The instructions do not specify any delimiters or warnings to ignore embedded instructions in the sheet data.
  • Capability inventory: The skill includes powerful write and delete capabilities, such as GOOGLESHEETS_BATCH_UPDATE and GOOGLESHEETS_DELETE_DIMENSION.
  • Sanitization: No sanitization or validation of the sheet content is mentioned before it is processed or used in further tool calls.
  • [DYNAMIC_EXECUTION]: The skill utilizes RUBE_SEARCH_TOOLS to dynamically fetch tool schemas at runtime. This means the exact behavior and interface of the tools are not statically defined in the skill and can be modified by the remote MCP server provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 12:40 PM
Security Audit — agent-trust-hub — googlesheets-automation