Automated Ticket Triage for HaloPSA with Gemini AI Summary Generation

Description

What it is

This workflow listens for new tickets in HaloPSA via webhook, generates a professional AI-powered summary of the issue using Gemini (or another LLM), and posts it back into the ticket as a private note.

It’s designed for MSPs using HaloPSA who want to reduce triage time and give engineers a clear head start on each support case.

⸻

✨ Features
• 🔔 Webhook trigger from HaloPSA on new ticket creation
• 🚧 Optional team filter (skip Sales or other queues)
• 📦 Extracts ticket subject, details, and ID
• 🧠 Builds a structured AI prompt with MSP context (NinjaOne, M365, CIPP)
• 🤖 Processes via Gemini or other LLM
• 📑 Cleans & parses JSON output (summary, next step, troubleshooting)
• 🧱 Generates a branded HTML private note (logo + styled sections)
• 🌐 Posts the note back into HaloPSA via API

⸻

🔧 Setup
Webhook
• Replace WEBHOOK_PATH and paste the generated Production URL into your HaloPSA webhook.
Guard filter (optional)
• Change teamName or teamId to skip tickets from specific queues.
Branding
• Replace YOUR_LOGO_URL and Your MSP Brand in the HTML note builder.
HaloPSA API
• In the HTTP node, replace YOUR_HALO_DOMAIN and add your Halo API token (Bearer auth).
LLM credentials
• Set your API key in the Gemini / OpenAI node credentials section.
(Optional) Adjust the AI prompt with your own tools or processes.

⸻

✅ Requirements
• HaloPSA account with API enabled
• Gemini / OpenAI (or other LLM) API key
• SMTP (optional) if you want to extend with notifications

⸻

⚡ Workflow overview

`🔔 Webhook → 🚧 Guard → 📦 Extract Ticket → 🧠 Build AI Prompt → 🤖 AI Agent (Gemini) → 📑 Parse JSON → 🧱 Build HTML Note → 🌐 Post to HaloPSA`