triage-meeting-prep

bởi microsoft

Chuẩn bị tóm tắt cuộc họp phân loại hàng tuần cho các vấn đề Needs-Triage của WinAppSDK. Sử dụng khi chuẩn bị cho các cuộc họp phân loại, xem xét các vấn đề Needs-Triage, tạo diff…

npx skills add https://github.com/microsoft/windowsappsdk --skill triage-meeting-prep

Triage Meeting Preparation

This skill helps prepare comprehensive summaries for the twice-weekly WinAppSDK triage meetings. It focuses on issues that actually need triage attention:

  1. Needs-Triage WITHOUT area label — These need area assignment (primary focus)
  2. Closed issues still labeled Needs-Triage — These need follow-up replies

Note: Open issues with both Needs-Triage AND an area-* label are excluded — they're already triaged and waiting for area team action.

Research-Backed Analysis

Every suggestion and reply is backed by research:

Research TypeWhat We SearchWhy
🔍 Similar IssuesGitHub issue searchFind duplicates, related bugs, past fixes
💻 Codebasegrep_search, semantic_searchUnderstand technical context, find workarounds
📚 Documentationfetch_webpage on MS docsVerify known limitations, expected behavior
📜 Git Historygit log, git blameIdentify area owners, recent changes

This ensures:

  • Suggested replies are technically accurate
  • Workarounds are validated against the codebase
  • We reference related issues customers may have missed
  • Area assignments are informed by code structure

Key Outputs: What's Changed Since Last Triage

The skill tracks four critical categories for each meeting:

CategoryDescriptionAction Required
🌟 Popular Issues≥5 reactions (community interest)May need priority attention
🆕 Created This WeekIssues created since last weekly triageReview + assign area
⏳ Older PendingOlder issues still without area labelsFollow up — why no action?
✅ Closed (need reply)Closed issues with customer follow-upDraft reply with [confidence:XX]

All suggestions include confidence scoring in [confidence:XX] format for easy filtering.

When to Use This Skill

Primary triggers — Use this skill when:

  1. Preparing for triage meetings — Generate a summary of issues needing area assignment
  2. Reviewing what's changed — See the diff of issues since the last triage run
  3. Drafting customer replies — Get research-backed suggested replies for closed issues
  4. Tracking pending items — Identify issues that still haven't been actioned from previous meetings

Anti-patterns — Do NOT use this skill when:

  • Reviewing a single specific issue (use Get-IssueDetails.ps1 for individual issues)
  • General issue searching without triage context
  • Non-WinAppSDK repositories

Prerequisites

  • GitHub CLI (gh) installed and authenticated (gh auth login)
  • Access to microsoft/WindowsAppSDK repository
  • Previous triage state file for diff comparison: Generated-Files/triageMeeting/previous-state.json

Migration note: If you have an existing Generated Files/triageMeeting/... directory from earlier runs, rename Generated Files to Generated-Files before using this skill, or pass explicit input/output paths to the scripts so you keep prior diff history.

Installing GitHub CLI

# Windows
winget install GitHub.cli

# Then authenticate
gh auth login

Step-by-Step Workflow

Workflow: Generate Triage Meeting Summary

  1. Load previous triage state from Generated-Files/triageMeeting/previous-state.json
  2. Fetch current Needs-Triage issues using the scripts:
    # Get all issues needing triage (no area label)
    ./.github/skills/triage-meeting-prep/scripts/Get-TriageIssues.ps1 -NoAreaOnly -OutputFormat summary
    
    # Or get JSON for processing
    $issues = ./.github/skills/triage-meeting-prep/scripts/Get-TriageIssues.ps1 -NoAreaOnly | ConvertFrom-Json
    
  3. Categorize issues by creation date:
    • 🆕 Created This Week — createdAt ≥ last weekly triage date
    • ⏳ Older Pending — createdAt < last weekly triage date
  4. Fetch area label definitions for classification:
    ./.github/skills/triage-meeting-prep/scripts/Get-RepositoryLabels.ps1 -Filter "area-*" -OutputFormat json
    
  5. For each no-area issue, analyze using Get-IssueDetails.ps1:
    • Run: ./Get-IssueDetails.ps1 -IssueNumber <number> -OutputFormat summary
    • Review the issue body, comments, and labels
    • Classify the area label by reasoning about the issue content against the area label definitions (see workflow for confidence rubric)
  6. Generate summary report with links to each issue's overview.md
  7. Save current state for next diff comparison

Important: Every issue without an area label gets a review. The summary links to each review file.

Available Scripts

The skill provides several PowerShell scripts for data fetching and analysis:

Get-TriageIssues.ps1

Fetches all issues with the Needs-Triage label from the repository.

# Get all Needs-Triage issues without area labels
./Get-TriageIssues.ps1 -NoAreaOnly -OutputFormat summary

# Get JSON output for programmatic processing
$issues = ./Get-TriageIssues.ps1 -NoAreaOnly | ConvertFrom-Json

Parameters:

  • -Repository — Repository in 'owner/repo' format (default: microsoft/WindowsAppSDK)
  • -State — Filter by state: 'open', 'closed', or 'all' (default: all)
  • -NoAreaOnly — Only include issues without area-* labels
  • -Limit — Maximum issues to fetch (default: 200)
  • -OutputFormat — Output format: 'json', 'summary', or 'table' (default: json)

Get-IssueDetails.ps1

Performs deep analysis of a single issue including scoring and recommendations.

# Analyze a specific issue
./Get-IssueDetails.ps1 -IssueNumber 1234 -OutputFormat summary

Get-RepositoryLabels.ps1

Fetches all label definitions from the repository including names, descriptions, and colors.

# Get all labels
./Get-RepositoryLabels.ps1 -OutputFormat summary

# Get only area labels
./Get-RepositoryLabels.ps1 -Filter "area-*" -OutputFormat table

# Get labels from a different repository
./Get-RepositoryLabels.ps1 -Repository "microsoft/terminal" -OutputFormat json

Parameters:

  • -Repository — Repository in 'owner/repo' format (default: microsoft/WindowsAppSDK)
  • -Filter — Filter pattern for label names (supports wildcards like 'area-*')
  • -Limit — Maximum labels to fetch (default: 500)
  • -OutputFormat — Output format: 'json', 'summary', or 'table' (default: json)

Use cases:

  • Understanding available area labels for issue categorization
  • Verifying label descriptions for triage decisions
  • Comparing label definitions across repositories
  • Finding labels that lack descriptions

Compare-TriageState.ps1

Compares current triage state with previous state to identify changes.

Save-TriageState.ps1

Saves the current triage state for future comparisons.

Output Structure

All outputs are saved to Generated-Files/triageMeeting/<date>/:

FilePurpose
summary.mdMain meeting document with tables linking to individual reviews
previous-state.jsonState snapshot for next diff (auto-generated)

Individual issue reviews are saved to Generated-Files/issueReview/<issue-number>/:

FilePurpose
overview.mdFull issue analysis with scores and suggested actions (linked from summary)
implementation-plan.mdTechnical implementation details (if applicable)

Key Output: The summary.md includes the Suggested Actions section extracted from each issue's overview.md, with [confidence:XX] scores.

Detailed Workflow

See workflow-triage-prep.md for the complete step-by-step process.

Issue Categories

CategoryDefinitionAction Needed
🌟 Popular Issues≥5 reactions (community interest)May need priority attention
🆕 Created This WeekcreatedAt is after last weekly triage dateUse Get-IssueDetails.ps1 + assign area
⏳ Older PendingCreated before this week, still no area labelFollow up — why not actioned?
✅ ResolvedIn previous state, NOT in currentAcknowledge (got area label or closed)
🏷️ Closed (cleanup)Closed + still has Needs-Triage labelRemove Needs-Triage label

Key distinction: "Created This Week" is based on when the issue was created (createdAt), not when it was first seen in triage runs. This ensures new issues are highlighted even if triage runs multiple times per week.

Summary.md Format

See template-summary.md for the full template structure.

Area Classification (LLM-Based)

Issues without an area-* label are classified by the agent using LLM reasoning — not keyword matching. The agent:

  1. Fetches area label definitions via Get-RepositoryLabels.ps1 -Filter "area-*"
  2. Reads the issue title, body, and comments from Get-IssueDetails.ps1
  3. Reasons about the best area match and assigns a confidence score

Confidence scale:

ScoreLevelMeaning
80–100HighIssue explicitly names the component; maps to exactly one area
60–79Medium-HighStrong signals (stack traces, API names) point to one area
40–59MediumReasonable match, but 2+ areas are plausible
20–39LowWeak signal; best guess
0–19Very LowNo clear signal; vague or multi-area

See workflow-triage-prep.md for the full confidence rubric and adjusters.

Action Item Types

ActionWhen to Suggest
Assign to milestoneHigh business importance, clear scope
Add area labelMissing component/area identification
Request clarificationRequirement clarity < 50
Close as duplicateSimilar closed issue found with fix
Close as by-designExpected behavior, not a bug
Needs investigationRepro unclear, needs debugging
Ready for workClear requirements, assignable

Troubleshooting

SymptomSolution
No previous state fileFirst run — all issues shown as "New"
gh CLI not installedRun winget install GitHub.cli then gh auth login
gh not authenticatedRun gh auth login and follow the prompts
Rate limitedWait and retry, or reduce --limit parameter
Missing issue detailsUse Get-IssueDetails.ps1 -IssueNumber <number>

Related Resources

Thêm skills từ microsoft

oss-growth
microsoft
Cá tính tăng trưởng OSS
agent-framework-azure-ai-py
microsoft
Xây dựng các tác nhân Azure AI Foundry bằng SDK Python của Microsoft Agent Framework (agent-framework-azure-ai). Sử dụng khi tạo các tác nhân bền vững với AzureAIAgentsProvider, sử dụng các công cụ được lưu trữ (trình thông dịch mã, tìm kiếm tệp, tìm kiếm web), tích hợp máy chủ MCP, quản lý chuỗi hội thoại hoặc triển khai phản hồi phát trực tuyến. Bao gồm các công cụ hàm, đầu ra có cấu trúc và các tác nhân đa công cụ.
development
airunway-aks-setup
microsoft
Thiết lập AI Runway trên AKS — từ cụm trống đến mô hình đang chạy. Bao gồm xác minh cụm, cài đặt controller, đánh giá GPU, thiết lập nhà cung cấp và triển khai đầu tiên. KHI NÀO: "thiết lập AI Runway", "onboard cụm AKS", "cài đặt AI Runway", "thiết lập airunway", "triển khai mô hình lên AKS", "suy luận GPU trên AKS", "thiết lập KAITO trên AKS", "chạy LLM trên AKS", "vLLM trên AKS", "thiết lập phục vụ mô hình trên AKS", "AI Runway controller".
devops
appinsights-instrumentation
microsoft
Hướng dẫn để instrument các ứng dụng web với Azure Application Insights. Cung cấp các mẫu telemetry, thiết lập SDK, và tài liệu tham khảo cấu hình. KHI NÀO: cách instrument ứng dụng, App Insights SDK, các mẫu telemetry, App Insights là gì, hướng dẫn Application Insights, ví dụ instrumentation, các phương pháp tốt nhất APM.
devops
applicationinsights-web-ts
microsoft
Instrument các ứng dụng trình duyệt/web bằng SDK JavaScript Application Insights (@microsoft/applicationinsights-web). Dùng cho Real User Monitoring (RUM) — lượt xem trang, nhấp chuột, phụ thuộc AJAX/fetch, ngoại lệ, sự kiện tùy chỉnh và dấu vết tác nhân GenAI phía trình duyệt tương quan với dấu vết OpenTelemetry phía backend. Bao gồm thiết lập SDK Loader Script và npm, tiện ích mở rộng framework (React, React Native, Angular), Click Analytics, trình khởi tạo telemetry và quy ước ngữ nghĩa OTel GenAI cho các span tác nhân/công cụ/mô hình phát ra từ trình duyệt.
devops
azure-ai-anomalydetector-java
microsoft
Xây dựng ứng dụng phát hiện bất thường với Azure AI Anomaly Detector SDK cho Java. Sử dụng khi triển khai phát hiện bất thường đơn biến/đa biến, phân tích chuỗi thời gian hoặc giám sát hỗ trợ AI.
development
azure-ai-language-conversations-py
microsoft
Triển khai Conversational Language Understanding (CLU) bằng SDK Python azure-ai-language-conversations. Sử dụng khi làm việc với ConversationAnalysisClient để phân tích ý định và thực thể trong hội thoại, xây dựng tính năng NLP, hoặc tích hợp hiểu ngôn ngữ vào ứng dụng.
development
azure-ai-ml-py
microsoft
Azure Machine Learning SDK v2 cho Python. Dùng cho không gian làm việc ML, công việc, mô hình, tập dữ liệu, tính toán và quy trình. Kích hoạt: "azure-ai-ml", "MLClient", "không gian làm việc", "đăng ký mô hình", "công việc đào tạo", "tập dữ liệu".
development