The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package globally. This is the official command-line tool for the Membrane platform, which is the provider of this skill. This tool is used to mediate all interactions with the Kern AI API.
[COMMAND_EXECUTION]: The skill utilizes the membrane CLI to perform actions such as searching for integrations, establishing connections, and executing API requests. These commands are restricted to the platform's ecosystem and do not involve arbitrary shell execution or system modification.
[CREDENTIALS_UNSAFE]: Security best practices are followed regarding credential handling. The skill explicitly advises against manual token management and instead uses Membrane's server-side connection system, which prevents sensitive API keys from being exposed in the local environment or agent logs.
[DATA_EXFILTRATION]: Network operations are performed via the membrane request proxy. This ensures that the agent communicates through an authenticated intermediary (Membrane) to the official Kern AI endpoints, reducing the risk of data being sent to unauthorized third-party domains.

kern-ai