sendblue

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The instructions utilize the membrane command-line interface to authenticate, manage connections, and execute API actions. These operations are restricted to the functionality provided by the Membrane platform.
  • [EXTERNAL_DOWNLOADS]: The skill includes instructions to install @membranehq/cli, which is the official tool provided by the vendor for interacting with their service. This is a standard dependency for the skill's operation.
  • [CREDENTIALS_UNSAFE]: The skill explicitly advises against manual credential handling, instructing the agent to use Membrane's connection system to manage authentication tokens securely on the server side.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 10:52 PM