billy

Pass

Audited by Gen Agent Trust Hub on May 2, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Membrane CLI (membrane) to manage connections and execute actions. These commands are part of the intended integration workflow.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli package from the official NPM registry. This is a legitimate vendor tool provided by the author to facilitate the integration.
  • [SAFE]: The skill explicitly instructs the agent to let Membrane handle credentials and never ask the user for API keys, which is a significant security best practice for managing sensitive tokens.
  • [SAFE]: No evidence of prompt injection, obfuscation, or unauthorized data exfiltration was found. The skill operations are transparent and consistent with its stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
May 2, 2026, 01:50 PM
Security Audit — agent-trust-hub — billy