makelog

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the @membranehq/cli for authentication, connection management, and executing actions. This includes commands like membrane login, membrane connect, and membrane action run.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli package from the official NPM registry, which is the standard method for accessing the vendor's tooling.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes the Membrane platform's core functionality to dynamically create and execute actions based on user-provided descriptions. This process is managed through the vendor's CLI and cloud infrastructure as part of the intended integration logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 07:38 PM