smartengage

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the official @membranehq/cli package from npm to facilitate integration tasks.
  • [COMMAND_EXECUTION]: Shell commands are used to manage connections and execute actions within the SmartEngage ecosystem.
  • [DATA_EXFILTRATION]: User data and parameters are transmitted to SmartEngage API endpoints through a proxy mechanism.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external data from SmartEngage.
  • Ingestion points: API response data from SmartEngage.
  • Boundary markers: None.
  • Capability inventory: Ability to perform write actions and proxy HTTP requests.
  • Sanitization: No data sanitization is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 11:19 PM