issue-triage

name	issue-triage
description	Use when triaging open GitHub issues for the PAS framework. Reads issues, classifies them, assesses priority, and produces a triage report for Discovery.

Issue Triage

Overview

Read open GitHub issues on the PAS repository, classify each one, assess priority, and produce a triage report that feeds into the Discovery phase alongside internal feedback signals.

When to Use

• At the start of a Discovery phase
• When the product owner asks about open issues or community requests

Process

1. Fetch open issues: Run gh issue list --repo ZoranSpirkovski/PAS --state open --json number,title,body,labels,createdAt,comments,author --limit 50
2. Classify each issue:
   • Bug: something is broken or behaving incorrectly
   • Feature request: a new capability or enhancement
   • Question: someone asking how to use PAS
   • Framework feedback: feedback about PAS itself routed from the self-evaluation system (target: framework:pas)
3. Assess priority:
   • HIGH: blocks users, data loss risk, or multiple people report the same thing
   • MEDIUM: degraded experience but workaround exists
   • LOW: nice-to-have, cosmetic, or single report
4. Check for duplicates: Group issues that describe the same underlying problem
5. Identify actionable items: Which issues have enough information to act on? Which need clarification?
6. Produce report: Write triage to workspace

Output Format

# Issue Triage Report

## Summary
- Open issues: {N}
- By type: {N} bugs, {N} feature requests, {N} questions, {N} framework feedback
- Needs clarification: {N}

## Actionable Issues

### #{number}: {title} ({type}, {priority})
**Author:** {author}
**Summary:** {one-sentence summary}
**PAS target:** {which artifact this relates to, if identifiable}
**Action:** {what addressing this would involve}

## Needs Clarification

### #{number}: {title}
**Missing:** {what information is needed}
**Suggested question:** {what to ask the author}

## Duplicates
{Groups of issues describing the same problem}

Quality Checks

• All open issues were read (none skipped)
• Classifications are based on issue content, not just title
• Priority reflects impact, not just recency
• Issues needing clarification have specific questions, not generic "please provide more info"

Common Mistakes

(Populated by feedback over time)