sheets
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the 'goog' CLI tool to interact with Google Sheets. It provides specific command patterns for creating spreadsheets, updating values, and modifying grid structures. The reliance on a local binary (target/debug/goog) is a standard method for tool-integrated skills.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it is designed to ingest and process data from external spreadsheets.\n
- Ingestion points: Data is read from Google Sheets via the
goog sheets values get-tablecommand and from local CSV/TSV files through the--dataargument.\n - Boundary markers: There are no instructions or delimiters provided to separate data content from agent instructions or to prevent the agent from following directives found within the ingested data.\n
- Capability inventory: The skill allows the agent to execute shell commands via the 'goog' tool and write data, including formulas, back to spreadsheets.\n
- Sanitization: The instructions do not specify any validation or sanitization protocols for data retrieved from remote or local files prior to its use by the agent.\n- [SAFE]: The skill's behavior is consistent with its stated purpose of spreadsheet management. It uses established authentication methods (goog auth login) and does not exhibit patterns of data exfiltration, credential theft, or malicious obfuscation.
Audit Metadata