gog
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
gogCLI tool to perform operations across Google Workspace services (Gmail, Drive, Sheets, etc.). All commands are aligned with the skill's stated purpose of automation and data management. - [PROMPT_INJECTION]: The skill processes untrusted data from external sources (e.g., email content, document text), creating an indirect prompt injection surface. This is mitigated by explicit safety instructions provided in the documentation.
- Ingestion points:
SKILL.md(viagmail get,docs cat,sheets getcommands). - Boundary markers: None explicitly defined in the provided command examples.
- Capability inventory: The
gogtool performs network operations (API calls) and file modifications (upload, write). - Sanitization: The skill recommends using the
--sanitize-contentflag for Gmail and--readonlyservice scopes to limit exposure.
Audit Metadata