The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli package globally via npm. This is an official tool provided by the platform vendor (membrane/membranedev) for managing integrations.
[COMMAND_EXECUTION]: The skill relies on shell commands to interact with the Membrane platform. These commands include membrane login, membrane connect, and membrane action run, which are standard operations for this CLI tool.
[CREDENTIALS_SAFE]: The skill explicitly advises against hardcoding or asking users for API keys. Instead, it uses the platform's connection system to handle authentication server-side, which is a recommended security practice.
[PROMPT_INJECTION]: The skill processes data from Cutt.ly (such as link statistics and shortened URLs) which represents a surface for indirect prompt injection.
Ingestion points: Data enters the context through the output of membrane action run (found in SKILL.md).
Boundary markers: None are explicitly defined in the provided prompt instructions for handling action output.
Capability inventory: The agent has the ability to execute shell commands using the Membrane CLI.
Sanitization: No sanitization or escaping of the action output is described before it is processed by the agent.

cuttly