The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation instructs the installation of the @membranehq/cli Node.js package. This is an official command-line interface provided by the vendor (membrane) to facilitate integrations.
[COMMAND_EXECUTION]: The skill utilizes several membrane CLI commands (e.g., membrane login, membrane action run, membrane request) to manage connections and execute API operations. These commands are the core mechanism for the skill's functionality.
[DATA_EXFILTRATION]: The skill provides a proxying capability via membrane request to interact with the LiveKit API. This allows the agent to communicate with external endpoints through the vendor's infrastructure, which is the intended design for secure authentication and request routing in this environment.
[PROMPT_INJECTION]: The skill acts as an integration that processes data from the LiveKit API, introducing a potential surface for indirect prompt injection from external data sources.
Ingestion points: Data enters the agent context through API responses from membrane action run and membrane request (SKILL.md).
Boundary markers: No specific boundary markers or instructions to ignore embedded commands in the API data are provided.
Capability inventory: The agent can execute shell commands via the membrane CLI and perform network operations through the vendor's proxy.
Sanitization: No explicit sanitization or validation of the external API content is mentioned.

livekit