The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli global package via NPM. This is a known vendor resource for the Membrane platform and is necessary for the skill's operations.
[COMMAND_EXECUTION]: Multiple shell commands are utilized to interact with the ZITADEL platform through the membrane utility. This includes logging in, searching for connectors, and executing API actions, which minimizes direct exposure of credentials by handling auth logic within the CLI.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes structured JSON data from external command outputs to determine subsequent logic.
Ingestion points: Output from membrane action list and membrane connection list commands (SKILL.md).
Boundary markers: No specific delimiters or instructions provided to the agent to disregard instructions potentially embedded in the tool's output.
Capability inventory: Execution of API actions via membrane action run and arbitrary network requests via membrane request.
Sanitization: No explicit sanitization or validation of the schema or descriptions returned by the CLI is described.

zitadel