Project: support ticket triage tool

You are an engineer at esc bash. The support inbox is overflowing and nobody has time to read every message just to route it. Your job is to build a tool that does the first pass automatically.

Set up the project

Run this to drop a folder of real-looking support tickets at /root/tickets/:

curl -fsSL https://raw.githubusercontent.com/Esc-Bash/project-init-scripts/main/ai/project-1/init.sh | bash

Have a look at the files it created so you know what you are working with.

What to build

Turn each raw ticket into a small, structured record a dashboard could load: what kind of issue it is, and how urgent it looks.

• Write a tool that reads every ticket in /root/tickets/ and, for each one, asks a model to classify it and return JSON with at least these two fields:
  • category - one of billing, technical, account, or other
  • priority - one of low, medium, or high
• Save each result to /root/triaged/ in a file named after the ticket, with a .json extension (so tickets/deploy-500.txt produces triaged/deploy-500.json).
• Every .json file must be valid JSON that jq can parse and must contain both fields.

This pulls together what you have practiced: a clear, specific prompt, asking for structured JSON output, and looping the call over many inputs from code. Keep the prompt strict so the model returns only JSON.

Press Submit once every ticket in /root/tickets/ has a matching, valid JSON result in /root/triaged/.

Your lab setup

This VM comes preconfigured with your AI keys - just type claude to get started. Your key and URL are already set as $OPENAI_BASE_URL and $OPENAI_API_KEY.