The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the membrane command-line interface to perform operations such as authentication, listing connections, and executing actions. These commands are necessary for the skill's documented functionality.
[EXTERNAL_DOWNLOADS]: The instructions include installing the @membranehq/cli package from npm. This is an official vendor resource used to facilitate the integration with the Membrane platform.
[CREDENTIALS_UNSAFE]: The skill explicitly advises against asking users for API keys or secrets, instead using a secure connection-based authentication flow managed by the Membrane platform.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface where user-provided descriptions are used to dynamically create actions via membrane action create. While this involves processing natural language to generate functionality, it is a core feature of the platform and does not exhibit malicious patterns in the static instructions.

fena