The Agent Skills Directory

[SAFE]: The skill facilitates interaction with the Novu API using the Membrane CLI. It follows security best practices by delegating credential management to a server-side service, avoiding the need for hardcoded secrets or local token storage.
[EXTERNAL_DOWNLOADS]: The skill references the installation of the @membranehq/cli Node.js package. This is an official utility provided by the skill's author for managing integrations.
[COMMAND_EXECUTION]: Shell commands are used to invoke the CLI for resource discovery and action execution. These commands are limited to the scope of the Membrane platform's functionality.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external JSON data from the Novu API to determine subsequent agent actions. This is an inherent part of the integration workflow and is documented as a low-risk capability.
Ingestion points: CLI output from membrane action list and membrane connection list (SKILL.md).
Boundary markers: None present.
Capability inventory: Subprocess execution via membrane action run and network requests via membrane request (SKILL.md).
Sanitization: No explicit sanitization or validation of the CLI output is described within the skill instructions.

novu