Skills
Navigate
skills/melvynx/ai-maxxing-skills/connect-google/Gen Agent Trust Hub

connect-google

Fail

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONPERSISTENCECOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the agent to download a bash script (gog-oauth.sh) from a remote GitHub repository and execute it locally using bash. This pattern of downloading and running remote code is high-risk.
  • [PERSISTENCE]: The instructions command the agent to modify the user's shell initialization files (~/.zshrc, ~/.zprofile, ~/.bashrc, and ~/.bash_profile) to export environment variables, ensuring specific configurations persist across all future shell sessions.
  • [COMMAND_EXECUTION]: The skill performs high-privilege operations including the use of brew trust on a third-party formula and npx to dynamically install additional skills from external sources.
  • [CREDENTIALS_UNSAFE]: The skill directs the agent to programmatically capture sensitive Google Cloud Platform credentials, specifically the client_id and client_secret, and write them to a local JSON file (gog-client.json) before importing them into the gog CLI tool.
Recommendations
  • HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/Melvynx/ai-maxxing-skills/main/skills/connect-google/scripts/gog-oauth.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Oct 23, 2026, 06:00 AM