cafe-24

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from the official NPM registry. This tool is the vendor's primary CLI for managing integrations.
  • [COMMAND_EXECUTION]: Uses local shell commands to interact with the Membrane platform via the membrane executable. These commands are used for session management, connector discovery, and action execution.
  • [SAFE]: Adheres to security best practices by utilizing a managed proxy for API requests, which ensures that sensitive authentication headers and credential refreshes are handled server-side rather than within the agent's prompt or local scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 04:39 PM