The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the membrane CLI to interact with the service. This is the intended way to use the platform and doesn't involve unsafe shell operations or privilege escalation.
[EXTERNAL_DOWNLOADS]: The instructions guide the user to install the @membranehq/cli package from npm. This is the vendor's official tool and is used to facilitate the integration.
[CREDENTIALS_UNSAFE]: The skill explicitly advises against hardcoding secrets and instead uses a secure login flow (membrane login) to manage authentication tokens server-side.
[PROMPT_INJECTION]: The skill includes commands that take natural language input as arguments (e.g., membrane action list --intent "QUERY"). This is a core feature for discovering API actions and is processed by the vendor's infrastructure rather than affecting the agent's internal safety logic.

jw-player