github-issue-triage

Triage GitHub issues by launching one background analysis task per issue with streaming results.

3|Updated Feb 7, 2026
One-click install
npx skills add https://github.com/gabrielnsmnto/kord-aios --skill github-issue-triage-gabrielnsmnto
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: github-issue-triage
Source: https://github.com/gabrielnsmnto/kord-aios/tree/main/.opencode/skills/github-issue-triage
Command: npx skills add https://github.com/gabrielnsmnto/kord-aios --skill github-issue-triage-gabrielnsmnto

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires typer, rich, and includes scripts (resource) components.

What problem does it solve? Manually reviewing a backlog of GitHub issues is slow and error-prone, especially when maintainers need to quickly identify critical bugs, closable duplicates, and unanswered questions across dozens of issues. ## Core Features & Use Cases - Parallel Issue Analysis: Launches one independent background task per issue so failures never cascade and results stream in real time. - Exhaustive Pagination: Uses the bundled gh_fetch.py script to fetch every issue and PR within a configurable time window via the GitHub CLI. - Structured Triage Report: Categorizes each issue as critical, closable, auto-respondable, needs-investigation, feature backlog, or needs-info, then generates a final comprehensive Markdown report with draft responses. - Use Case: A maintainer returning after a weekend runs the triage to instantly see which of 40 new issues are critical bugs, which can be closed, and which already have draft replies ready. ## Quick Start Triage all GitHub issues from the last 48 hours in this repository and stream the analysis results as each issue completes.

Frequently Asked Questions about github-issue-triage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I triage GitHub issues automatically with an AI agent?▼

Run the triage workflow which fetches all issues via the GitHub CLI, launches one background analysis task per issue, and streams categorized results as each completes. It ends with a comprehensive report grouping issues by criticality and recommended action.

How to fetch all GitHub issues with pagination using gh CLI?▼

Use the bundled gh_fetch.py script with commands like ./gh_fetch.py issues --hours 48 --output json. It paginates in batches of 500 using created-date search filters and deduplicates results until no more pages exist.

Does this issue triage require the GitHub CLI?▼

Yes, the workflow depends on the gh CLI being installed and authenticated, since both the fetch script and the per-issue comment retrieval use gh commands like gh issue list and gh issue view.

Why process each GitHub issue as a separate background task?▼

Per-issue background tasks provide isolation so one failed analysis does not affect others, enable parallel execution for speed, and allow results to stream to the user immediately as each task completes instead of waiting for the full batch.

What are the limitations of automated GitHub issue triage?▼

The analysis depends on issue body quality and available PR data, so vague issues may be classified as unclear or needing info. Pagination is capped at 20 pages as a safety limit, and draft responses flagged NEEDS_MANUAL_REVIEW still require maintainer judgment.