The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the user/agent to install and execute the @membranehq/cli package. This is a legitimate tool provided by the skill's author (membranedev) for interacting with their platform.
[CREDENTIALS_UNSAFE]: The skill correctly avoids requesting or hardcoding API keys. It uses membrane login and membrane connect to manage OAuth flows and server-side credential storage, which is a recommended security practice.
[PROMPT_INJECTION]: There is a minor indirect prompt injection surface where the skill passes natural language strings to the CLI via the --intent and create arguments. While this data is processed by the Membrane platform to find or build actions, the risk is localized to the platform's interpretation of those strings.

gopay