What problem does it solve?
It reduces the backlog and confusion of open GitHub issues by automatically categorizing new reports, identifying likely duplicates, and closing redundant threads.
Core Features & Use Cases
- Duplicate detection workflow: Compares unlabeled issues against existing open and closed issues using title keyword search, then closes confirmed duplicates with a comment pointing to the original.
- Deterministic labeling: Applies consistent labels based on issue signals (bug, enhancement, question, needs-info), with a fallback for ambiguous reports.
- Needs-information follow-up: Posts a short, targeted request for the specific details required to reproduce or understand the issue.
- Operational summary: Produces a concise triage report counting how many issues were closed as duplicates vs. labeled by category.
Quick Start
Ask the AI to triage issues and label/close duplicates for the repository’s open unlabeled GitHub issues using the GitHub CLI.