The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or high-risk security vulnerabilities were detected in the skill instructions.
[EXTERNAL_DOWNLOADS]: The skill recommends installing the @membranehq/cli package from the npm registry. This is a legitimate tool provided by the vendor (membranedev) for interacting with their platform.
[COMMAND_EXECUTION]: The skill utilizes the membrane command-line interface to manage connections and execute actions. These operations are scoped to the Membrane platform's managed authentication and integration system and do not involve arbitrary shell execution.
[CREDENTIALS_UNSAFE]: The skill demonstrates secure credential management by advising against the use of raw API keys and instead leveraging Membrane's managed connection system for authentication, which handles credential lifecycles server-side.

repuso