google-vertex-ai

Pass

Audited by Gen Agent Trust Hub on May 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from the npm registry. This is a vendor-provided tool necessary for the skill's functionality.
  • [COMMAND_EXECUTION]: The skill utilizes several CLI commands (e.g., membrane login, membrane connect, membrane action run) to manage resources and execute tasks within Google Vertex AI.
  • [SAFE]: The skill implements strong security practices by utilizing a centralized connection manager. It explicitly instructs the agent not to request or handle raw API keys, reducing the risk of credential exposure.
  • [COMMAND_EXECUTION]: The skill features dynamic action creation via membrane action create, which generates integration logic based on natural language descriptions. While this is a core feature of the platform, it introduces a potential indirect prompt injection surface if the agent passes unvalidated user input into the action description.
Audit Metadata
Risk Level
SAFE
Analyzed
May 1, 2026, 03:09 PM