skills/usetemi/skills/google-drive/Gen Agent Trust Hub

google-drive

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates file operations by executing the rclone CLI via the subprocess module. Arguments are passed as lists to prevent shell injection vulnerabilities. This functionality is essential for the skill's operation as a transport layer for Google Drive.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies Google Drive as a source of untrusted data. It includes mandatory instructions for the agent to summarize pulled documents and seek user confirmation before processing them, specifically to defend against malicious instructions embedded in third-party files.
  • [EXTERNAL_DOWNLOADS]: The setup instructions reference external resources, including the rclone and uv binaries and supplementary skills from the official Anthropics repository. These downloads and references target well-known and trusted services for legitimate configuration and extension purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 03:21 AM
Security Audit — agent-trust-hub — google-drive