issue-triage

Categorize, prioritize, and respond to GitHub issues using labels and response templates.

Updated Apr 20, 2020
One-click install
npx skills add https://github.com/UnterrainerInformatik/java-rdb-utils --skill issue-triage-unterrainerinformatik
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: issue-triage
Source: https://github.com/UnterrainerInformatik/java-rdb-utils/tree/main/.agents/skills/issue-triage
Command: npx skills add https://github.com/UnterrainerInformatik/java-rdb-utils --skill issue-triage-unterrainerinformatik

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Maintaining a healthy GitHub issue backlog requires repetitive manual work: reading each issue, deciding its category, assigning priority labels, and writing responses. This Skill automates that triage workflow so maintainers can process backlogs in minutes instead of hours. ## Core Features & Use Cases - Issue Categorization: Classifies issues as bug reports, feature requests, questions, duplicates, or invalid based on content indicators like stack traces and reproduction steps. - Priority Assessment: Assigns P0-P3 priority levels (critical, high, medium, low) based on impact criteria such as security risk, data loss, and user reach. - Response Templates: Provides ready-made Markdown responses for common scenarios including requests for more information, duplicates, acknowledged bugs, and answered questions. - Use Case: After returning from vacation, ask the agent to triage the last 15 open issues; it fetches them via GitHub MCP or the gh CLI, labels each one, suggests responses, and produces a summary report. ## Quick Start Triage the last 15 open issues in my repository, categorize and prioritize them, then suggest responses for each.

Frequently Asked Questions about issue-triage

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

FAQPage Schema
How do I triage GitHub issues automatically?▼

Fetch open issues with the GitHub MCP list_issues tool or the gh issue list command, then categorize each by content indicators such as stack traces for bugs or feature phrasing for enhancements. Apply labels, assign a P0-P3 priority, and respond using the provided templates.

What tool should I use to fetch GitHub issues, MCP or gh CLI?▼

The GitHub MCP server is recommended because it uses fewer tokens through structured calls like list_issues and add_labels. The gh CLI works as an alternative with commands like gh issue list --json, but it is less token-efficient.

How are issue priorities decided during triage?▼

Priorities follow four levels: critical for security vulnerabilities or data loss, high for broken core features or regressions, medium for edge cases and valuable enhancements, and low for cosmetic or nice-to-have items. Each level maps to a label and milestone action.

Can this handle duplicate GitHub issues?▼

Yes, it searches for similar issues using GitHub keyword search and recently closed issues before marking duplicates. It then links to the original issue, closes the duplicate with a polite comment, and directs reporters to the original thread.

How do I reduce token usage when triaging many issues?▼

Batch multiple issues into a single prompt instead of processing one by one, cache the fetched issue list for follow-up commands, and read only the first post plus recent comments rather than entire threads. These strategies cut token usage by roughly 40-60 percent.