The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the @membranehq/cli package from the official NPM registry to facilitate communication with the Snowflake integration service.
[COMMAND_EXECUTION]: The skill utilizes local shell commands (membrane login, membrane connection, membrane action, membrane request) to manage authentication, discover available Snowflake actions, and execute data operations.
[INDIRECT_PROMPT_INJECTION]: The skill processes structured and unstructured data returned from the Snowflake API, which represents a potential surface for indirect prompt injection if the stored data contains malicious instructions intended for the agent.
Ingestion points: Data retrieved from Snowflake via membrane action run and membrane request (SKILL.md).
Boundary markers: None explicitly defined in the skill instructions to delimit external data from agent commands.
Capability inventory: The agent can execute system commands through the Membrane CLI and perform network requests via the Membrane proxy (SKILL.md).
Sanitization: The skill relies on the underlying agent's default safety guardrails and Membrane's server-side schema validation.

snowflake