novu

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the official Membrane CLI (@membranehq/cli) globally via NPM. This is a standard requirement for using the vendor's integration platform.
  • [COMMAND_EXECUTION]: Uses the membrane CLI to perform administrative and operational tasks, including authentication, connection management, and action execution. These commands are localized to the platform's ecosystem.
  • [PROMPT_INJECTION]: The skill includes instructions that process natural language inputs (intents) to discover or generate notification logic, creating a surface for indirect prompt injection.
  • Ingestion points: SKILL.md (lines 64, 75) uses user-supplied intent and description strings in CLI commands.
  • Boundary markers: Absent in the command-line interpolation.
  • Capability inventory: The CLI has the capability to run actions (membrane action run) and create new logic (membrane action create).
  • Sanitization: Relies on the backend Membrane platform to validate and sanitize the natural language intents before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 10:53 PM
Security Audit — agent-trust-hub — novu