skills/openai/plugins/google-sheets/Gen Agent Trust Hub

google-sheets

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Indirect Prompt Injection Surface: The skill is designed to read and process content from external Google Sheets, which represents a potential surface for indirect prompt injection. * Ingestion points: Data is ingested via tools such as get_cells, get_range, and search_spreadsheet_rows as described in references/reference-live-read-search-safety.md and references/reference-edit-workflow.md. * Capability inventory: The agent has the ability to perform complex writes and modifications using batch_update_spreadsheet, addTable, and addChart. * Boundary markers: The skill includes instructions in SKILL.md to treat provided sheets as read-only unless explicitly asked to edit, providing a logical boundary. * Sanitization: There are no explicit sanitization or filtering steps defined for the content read from cells before it is processed by the agent.
  • External Documentation References: The skill includes links to official Google Workspace and Google Sheets API documentation. These references provide technical guidance for formula syntax, chart recipes, and batch update request structures.
  • Platform Plugin Integration: The skill utilizes an internal spreadsheet plugin and specific MCP tools for workbook creation and management, which is consistent with the skill's intended purpose and authorship.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 11:39 AM
Security Audit — agent-trust-hub — google-sheets