AI Assistant

The AI Assistant leverages large language models to help support teams work efficiently.

Supported AI Providers

• OpenAI — Standard OpenAI API using the latest available model.
• Azure OpenAI — Enterprise Azure-hosted OpenAI and other models.
• Gemini — Google's AI model.

Setup & Configuration

Navigate to Admin > AI features to configure desired features.

Custom Rules: A Custom Rules field allows providing additional instructions the AI will follow. These are added to the system prompt. Example rules: "Always be polite and professional", "Our business hours are 9 AM - 5 PM EST", "Never promise refunds without manager approval".

Knowledge Base Integration: Enable "Include KB in AI responses" to allow the AI to automatically search your knowledge base when answering questions. The AI will use vector similarity search (SaaS) or keyword search to reference specific articles.

Core AI Features

AI-Generated Ticket Replies

The most common use case. Open a ticket, click the AI Assistant button, select "Create a reply to this ticket". AI analyzes the entire conversation and generates a response considering: customer's question, all follow-up messages, KB articles, canned responses, custom rules, and the conversation language.

Conversation Summaries

Click AI Assistant > "Summarize the conversation". AI extracts factual information: technical issues, current system state, actions taken, error messages, system configurations. Results are chronologically ordered.

Response Improvement

Write a draft response, then click AI Assistant > "Enhance your draft response". AI fixes grammar, improves readability, and expands brief points while preserving tone and meaning.

KB Article Generation

Open a resolved ticket, click AI Assistant > "Generate knowledge base content". AI generates article subject, body (HTML), and screenshot suggestions. Click "Proceed to article editor" to create the KB article.

Custom AI Chat

Click AI Assistant > "Custom AI-chat..." for open-ended conversation about any ticket. Context is preserved for multi-turn conversations. Example uses: "What are the main pain points?", "Suggest three possible solutions", "Search KB for this".

AI-Powered Automation Rules

Suggest Ticket Category Using AI

Automatically sets the most appropriate ticket category based on content analysis. Trigger: "When ticket is created", Action: "Suggest ticket category using AI".

Generate Reply with AI

Automatically generates and posts a response. Configuration options include making the reply visible to technicians only. Use cases: auto-respond outside business hours, generate first responses, provide initial troubleshooting steps.

Built-in AI Tools

• Knowledge Base Search (query_docs) — Automatically searches KB and documentation when needed.
• Canned Responses Search (search_canned_responses) — Searches template responses for relevant content.
• Image Description (describe_image) — Analyzes images and screenshots attached to tickets (Beta).

External Tools

Extend the AI Assistant with custom API integrations. Create HTTP endpoints that the AI can call automatically to retrieve data or perform actions.

Each external tool needs: Name (e.g. check_order_status), Description (tells AI when to use it), URL (POST endpoint), Headers (JSON, e.g. Authorization), and Parameters (name, description, required flag).

API contract: POST with JSON body, returns JSON with result or error field. Timeout: 3 minutes maximum.