The Agent Skills Directory

[SAFE]: No malicious patterns, such as obfuscation, persistence mechanisms, or unauthorized data exfiltration, were detected during the analysis.
[EXTERNAL_DOWNLOADS]: The skill instructs users to install the @membranehq/cli package via NPM. This is a legitimate dependency from the official organization associated with the skill's author (membranedev) and is used for its intended purpose.
[COMMAND_EXECUTION]: Utilizes the membrane CLI to perform operations such as service authentication, connection management, and triggering payment actions. These commands are necessary for the integration's functionality and do not exhibit suspicious behavior.
[CREDENTIALS_UNSAFE]: Follows security best practices by explicitly instructing the agent and user to use platform-managed connections rather than manual API keys or tokens, reducing the risk of credential exposure.

cashfree