Company logoOpen Ticket AI

Open Ticket Automation

Integration layer for OTOBO, Znuny/OTRS, and Zammad

Open Ticket Automation

Welcome to the Open Ticket Automation (OTAI) documentation.

OTAI Runtime is the core execution engine that bridges your ticketing system with custom AI models. It is designed for security-first, on-premise inference, ensuring that your sensitive ticket data never leaves your infrastructure.

What is OTAI Runtime doing?

The Runtime acts as a secure containerized environment (Docker-based) that handles the lifecycle of ticket automation:

  1. Ticket Ingestion: Connects to systems like OTOBO, Znuny/OTRS, or Zammad via dedicated connector plugins.
  2. Local Inference: Uses a custom-trained model (built specifically for your environment from your QueueSpec metadata) to classify, prioritize, or summarize tickets.
  3. Action Execution: Automatically routes tickets, sets fields, or drafts responses directly in your ticketing system based on model results.
  4. Full Audit Trail: Logs every decision and action locally for compliance and quality control.

Key Principles

  • Privacy First: No ticket content is ever sent to the cloud. Only the model artifacts (weights) are downloaded during setup.
  • Customization: Each customer gets a model trained on their specific taxonomy (QueueSpec), not a generic one.
  • Low Latency: Local execution means no round-trips to external APIs.
  • Auditability: Every AI suggestion can be traced back to the specific version of the model used.