The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the membrane CLI to manage connections, discover actions, and execute API calls. This is a standard and intended use of the platform's tooling.
[EXTERNAL_DOWNLOADS]: Recommends the installation of the @membranehq/cli package from the official npm registry. As this is a vendor-owned resource (membranedev/membranehq), it is considered a safe and expected dependency for the skill's functionality.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from an external CRM (Kustomer).
Ingestion points: Data retrieved via membrane action run (e.g., customer details, conversation messages, notes) enters the agent's context.
Boundary markers: None explicitly provided in the skill instructions to delimit external data from agent instructions.
Capability inventory: The skill has the capability to write data (create/update customers and conversations) and execute platform-generated actions via the membrane CLI.
Sanitization: The skill relies on the underlying agent's safety protocols and the Membrane platform's execution environment for sanitization.

kustomer