Ticket Auto-Triage

Goal

Automate issue labeling for unlabeled GitHub issues

How to Run

Run this loop in a GitHub repository to automatically label unlabeled issues using AI analysis.

1. 01

   Initiate the Loop

   Execute the kickoff prompt in your preferred coding agent environment (Cursor, Claude Code, Codex, etc.)

2. 02

   Monitor Progress

   Observe the agent's labeling decisions and verify accuracy as it processes each issue

3. 03

   Review Results

   After loop completion, audit labeled issues to ensure proper categorization

Workflow Steps

1. 01

   Check for Unlabeled Issues

   Run 'gh issue list' to identify open issues missing labels

2. 02

   Analyze Issue Content

   Review title, body, and comments to determine appropriate labels (bug, feature, documentation, etc.)

3. 03

   Apply Labels

   Add relevant labels to the issue using 'gh issue edit'

4. 04

   Verify Changes

   Confirm labels were successfully applied and issue is now categorized

5. 05

   Repeat Process

   Continue until no unlabeled issues remain or max iterations reached

Start the "Ticket Auto-Triage" loop.

Goal: Automate issue labeling for unlabeled GitHub issues
Max iterations: 10
Between iterations run: gh issue list --state open --label ''
Exit when: No unlabeled issues remain


Start the Ticket Auto-Triage loop to automatically label unlabeled issues in this repository. Begin by checking for open issues without labels, then analyze each one and apply appropriate labels based on content.

Self-pace this loop. After each iteration, run `gh issue list --state open --label ''` and evaluate the output, and only continue if the exit condition is not met (No unlabeled issues remain). Stop when the exit condition passes or 10 iterations are reached. Give a short status update each pass.