The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli package from npm and utilizes npx to run the latest version. This package is an official tool provided by the vendor 'membranedev'.
[COMMAND_EXECUTION]: The skill relies on several shell commands executed via the Membrane CLI, including membrane login, membrane action list, membrane action run, and membrane request, to perform its primary functions.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes dynamic data from the Raet/Membrane API to discover and execute actions.
Ingestion points: Data returned from membrane action list (action IDs and input schemas) and membrane connection list (connection IDs).
Boundary markers: None; there are no instructions provided to the agent to treat fetched action descriptions or schemas as untrusted content.
Capability inventory: The skill can execute arbitrary actions (membrane action run) and make network requests (membrane request) based on the fetched data.
Sanitization: There is no evidence of sanitization or validation performed on the action definitions before they are used to generate CLI commands.

raet