scrapingdog

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructs the user to install @membranehq/cli, which is the official command-line interface provided by the vendor (membranedev).
  • [SAFE]: The skill emphasizes security by instructing the agent to never ask users for API keys, instead relying on Membrane's server-side connection management for authentication.
  • [SAFE]: All documented operations (searching for connectors, creating connections, and running actions) are standard functionalities of the Membrane ecosystem for managing third-party integrations.
  • [SAFE]: No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected in the instructions or command examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 07:11 PM