quality-gates

作者: microsoft

在每个工作流阶段运行质量门验证。支持4种门类型:规范质量、规范到计划、计划到任务、完整性。生成门通过/失败…

npx skills add https://github.com/microsoft/github-copilot-modernization --skill quality-gates

User Input

You MUST consider the user input before proceeding (if not empty).

Output

Reports and checkpoints are written under your task's Artifact path: (from task metadata).

checklists/requirements.md           ← spec-quality gate
checkpoints/spec-to-plan.yaml        ← spec-to-plan gate
traceability-matrix.yaml             ← completeness gate
migration-summary.md                 ← completeness gate

Input

type = spec-quality | spec-to-plan | plan-to-tasks | completeness

Workflow

Step 1: Load Context

Load from cwd:

  • constitution.md — non-negotiable; violations are always CRITICAL
  • spec.md, plan.md, tasks.md, checkpoints/ (as relevant for the gate type)

Step 2: Gate Routing

Load the checklist reference for the selected gate:

typeReferenceKey Output
spec-qualityreferences/gate-spec-quality.mdchecklists/requirements.md
spec-to-planreferences/gate-spec-to-plan.mdcheckpoints/spec-to-plan.yaml
plan-to-tasksreferences/gate-plan-to-tasks.mdcheckpoints/plan-to-tasks.yaml
completenessreferences/gate-completeness.mdtraceability-matrix.yaml + migration-summary.md

Step 3: Execute Gate

Follow the process defined in the referenced gate file.

Step 4: Verdict

  • PASS: all checklist items pass → ✓ [gate] PASSED.
  • FAIL: any CRITICAL → ✗ [gate] FAILED. N critical issues. + remediation steps

Rules

  • READ-ONLY: never modify source code; only write checklists, checkpoints, reports.
  • Max 50 findings per gate.
  • Deterministic: rerunning without changes = same results.
  • Never hallucinate: if artifact is missing, report it.

Severity

LevelMeaning
CRITICALCheckpoint failed, broken traceability, constitution violation, requirement not implemented, testing strategy deviation without documented blocker
HIGHPartial requirement, implementation drift, untested acceptance criteria
MEDIUMIncomplete non-functional coverage, terminology drift
LOWStyle/naming, documentation gaps

Resources

  • references/gate-spec-quality.md — spec quality checklist and process
  • references/gate-spec-to-plan.md — spec-to-plan checklist and process
  • references/gate-plan-to-tasks.md — plan-to-tasks checklist and process
  • references/gate-completeness.md — completeness checklist and process (includes plan coverage)
  • templates/report-template.md — migration summary structure

来自 microsoft 的更多技能

oss-growth
microsoft
OSS增长黑客角色
agent-framework-azure-ai-py
microsoft
使用Microsoft Agent Framework Python SDK(agent-framework-azure-ai)构建Azure AI Foundry代理。在创建使用AzureAIAgentsProvider的持久化代理、使用托管工具(代码解释器、文件搜索、网络搜索)、集成MCP服务器、管理对话线程或实现流式响应时使用。涵盖函数工具、结构化输出和多工具代理。
development
airunway-aks-setup
microsoft
在AKS上设置AI Runway——从裸集群到运行模型。涵盖集群验证、控制器安装、GPU评估、提供商设置和首次部署。适用场景:“设置AI Runway”、“接入AKS集群”、“安装AI Runway”、“airunway设置”、“将模型部署到AKS”、“在AKS上进行GPU推理”、“在AKS上配置KAITO”、“在AKS上运行LLM”、“在AKS上使用vLLM”、“在AKS上设置模型服务”、“AI Runway控制器”。
devops
appinsights-instrumentation
microsoft
使用Azure Application Insights对Web应用进行插桩的指南。提供遥测模式、SDK设置和配置参考。适用场景:如何对应用进行插桩、App Insights SDK、遥测模式、什么是App Insights、Application Insights指南、插桩示例、APM最佳实践。
devops
applicationinsights-web-ts
microsoft
使用Application Insights JavaScript SDK(@microsoft/applicationinsights-web)为浏览器/Web应用添加检测。用于真实用户监控(RUM)——页面视图、点击、AJAX/fetch依赖项、异常、自定义事件,以及与后端OpenTelemetry追踪关联的浏览器端GenAI代理追踪。涵盖SDK加载器脚本和npm设置、框架扩展(React、React Native、Angular)、点击分析、遥测初始化器,以及从浏览器发出的代理/工具/模型跨度所遵循的OTel GenAI语义约定。
devops
azure-ai-anomalydetector-java
microsoft
使用适用于 Java 的 Azure AI 异常检测器 SDK 构建异常检测应用程序。在实现单变量/多变量异常检测、时间序列分析或 AI 驱动的监控时使用。
development
azure-ai-language-conversations-py
microsoft
使用azure-ai-language-conversations Python SDK实现对话语言理解(CLU)。当使用ConversationAnalysisClient分析对话意图和实体、构建NLP功能或将语言理解集成到应用程序中时使用。
development
azure-ai-ml-py
microsoft
Azure Machine Learning SDK v2 for Python。用于机器学习工作区、作业、模型、数据集、计算资源和管道。 触发词:“azure-ai-ml”、“MLClient”、“工作区”、“模型注册表”、“训练作业”、“数据集”。
development