The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the @membranehq/cli package from npm, which is the official tool provided by the vendor (Membrane) for managing service integrations.
[COMMAND_EXECUTION]: The skill utilizes the membrane CLI for various tasks, including logging in, connecting to services, and executing actions. These commands run in the local shell to interact with the Membrane platform.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from an external source (Graphy).
Ingestion points: Data returned from commands like membrane action list and membrane action run (e.g., user data, order details) is brought into the agent context.
Boundary markers: The skill does not employ specific boundary markers or instructions to isolate or ignore potentially malicious content within the external data.
Capability inventory: The skill has capabilities for executing shell commands and creating new automated actions, representing a significant capability surface.
Sanitization: No specific sanitization or validation of external data is described in the provided instructions.

graphy