The Agent Skills Directory

[SAFE]: The skill follows best practices by using a managed CLI (Membrane) to handle authentication and credentials, preventing the exposure of API keys or secrets.
[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli package from the official NPM registry. This is a verified vendor resource owned by membranedev.
[COMMAND_EXECUTION]: The skill utilizes the membrane command-line tool to perform actions like searching for connectors, logging in, and executing API requests. These operations are restricted to the intended functionality of the integration.
[INDIRECT_PROMPT_INJECTION]: The skill processes data returned from the Everestcard API via membrane action run and membrane request. This creates an ingestion surface for untrusted external data.
Ingestion points: API responses from membrane request and action outputs.
Boundary markers: None explicitly defined in the instructions.
Capability inventory: Subprocess execution (via membrane CLI) and network operations.
Sanitization: Standard LLM-side sanitization is assumed; no specific escaping is implemented in the skill scripts.

everestcard