The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the @membranehq/cli tool for platform operations. This is a vendor-owned resource (membranedev) used to manage authentication and interact with the Membrane API server-side.
[EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from the npm registry. This is a standard installation of a trusted developer tool required for the skill's functionality.
[DYNAMIC_EXECUTION]: The skill supports the dynamic creation of actions via the membrane action create command. This process involves the Membrane platform building automation logic based on natural language descriptions provided by the user.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8).
Ingestion points: Data enters the agent context through natural language strings provided in the --intent and --description flags, as well as the output from membrane action run.
Boundary markers: No explicit delimiters are used in the documentation for shell command arguments.
Capability inventory: The skill performs shell execution via the membrane CLI and npm package manager.
Sanitization: No sanitization or validation of the input strings or action outputs is specified in the instructions.

rd-station