gogcli

Warn

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The instructions direct the user to install the gogcli binary from a personal GitHub repository (github.com/steipete/gogcli) via Homebrew, AUR, or by building from source. This source is not associated with a known trusted organization or established service provider.
  • [COMMAND_EXECUTION]: The skill's core functionality relies on executing the gog binary to perform sensitive operations across multiple Google Workspace services, including Gmail, Drive, and Admin SDK.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its integration with external data sources.
  • Ingestion points: The skill retrieves data from Gmail messages (gog gmail get), Google Drive files (gog drive download), Google Docs (gog docs cat), and Google Sheets (gog sheets get).
  • Boundary markers: There are no instructions for the agent to use delimiters or ignore embedded instructions when processing retrieved content.
  • Capability inventory: The agent has high-privilege capabilities including sending emails (gog gmail send), deleting files (gog drive delete), and managing Workspace users (gog admin users suspend).
  • Sanitization: There is no requirement or guidance for the agent to sanitize or validate the content retrieved from external sources before processing it.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 4, 2026, 01:17 PM