The Agent Skills Directory

[SAFE]: The skill utilizes the @membranehq/cli Node.js package, which is an official tool provided by the vendor. This is a standard and safe method for facilitating the described integration.\n- [SAFE]: Authentication is managed via a secure browser-based OAuth flow (membrane login), which prevents the exposure of sensitive credentials within the skill's instructions or the agent's environment.\n- [SAFE]: The skill includes explicit instructions for the AI agent to avoid requesting API keys or tokens from the user, promoting secure credential management practices.\n- [SAFE]: All network operations and API calls are conducted through the vendor's managed CLI tool, ensuring that data is handled according to the platform's security protocols.\n- [SAFE]: No malicious patterns, such as prompt injection, obfuscation, or unauthorized data exfiltration, were identified during the analysis.

gladly