claid-ai

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes the vendor's official CLI package (@membranehq/cli) for all operations, which is a standard and safe integration practice.
  • [SAFE]: Authentication is handled via the platform's secure login flow (membrane login), avoiding the need for hardcoded secrets or requesting sensitive API keys from the user.
  • [SAFE]: Network interactions with the Claid AI API are routed through a managed proxy that handles credential injection and refresh server-side, minimizing the risk of credential exposure in local logs or prompts.
  • [SAFE]: The instructions explicitly include security best practices, such as letting the platform manage credentials and using pre-built actions to ensure secure communication with external services.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 03:03 PM