nimble-crm

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing the @membranehq/cli package from npm. This is the official tool provided by the vendor for managing integrations.
  • [COMMAND_EXECUTION]: All functionality is accessed via the membrane command-line tool. This includes logging in, connecting to Nimble, and running specific CRM actions.
  • [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because it processes data retrieved from external CRM records.
  • Ingestion points: Data returned from membrane action run is intended to be processed and interpreted by the agent.
  • Boundary markers: The instructions do not define specific delimiters or boundary markers for the data returned from CRM actions.
  • Capability inventory: The skill includes capabilities to search, create, and execute actions via the CLI.
  • Sanitization: No automated sanitization of the CRM data is mentioned before it reaches the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 09:11 PM
Security Audit — agent-trust-hub — nimble-crm