The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the Membrane CLI via npm (@membranehq/cli). This is a legitimate dependency provided by the skill vendor for interacting with their platform.
[COMMAND_EXECUTION]: The skill utilizes several membrane CLI commands for authentication (membrane login), connection management (membrane connect), and action discovery. These are standard operations for the platform's workflow and do not involve unauthorized privilege escalation.
[REMOTE_CODE_EXECUTION]: The skill facilitates dynamic action generation and execution through membrane action create and membrane action run. This pattern allows the agent to define and execute logic on the Membrane server. While this represents a form of remote execution, it is an intended core feature of the Membrane integration and is performed within the user's authenticated platform session.
[DATA_EXFILTRATION]: The skill follows security best practices by delegating credential management to the Membrane platform. It explicitly instructs against asking users for API keys, thereby preventing the local exposure of sensitive secrets within the agent's context.

seatable