Agent Skills › Azure/agent-landing-zone › iac-validation

iac-validation

GitHub

为 AI Landing Zone 资产构建基于证据的验证计划，涵盖 Bicep、脚本及 Azure 部署行为。适用于实现后或回归调查期间，通过映射验收标准、运行测试和预检来评估风险并报告结果。

infra/.github/skills/iac-validation/SKILL.md Azure/agent-landing-zone

触发场景

AI Landing Zone 资产实现完成后进行回归调查时

安装

npx skills add Azure/agent-landing-zone --skill iac-validation -g -y

更多选项

非标准路径

npx skills add https://github.com/Azure/agent-landing-zone/tree/main/infra/.github/skills/iac-validation -g -y

不安装直接使用

npx skills use Azure/agent-landing-zone@iac-validation

指定 Agent (Claude Code)

npx skills add Azure/agent-landing-zone --skill iac-validation -a claude-code -g -y

安装 repo 全部 skill

npx skills add Azure/agent-landing-zone --all -g -y

预览 repo 内 skill

npx skills add Azure/agent-landing-zone --list

SKILL.md

Frontmatter

{
    "name": "iac-validation",
    "description": "Builds an evidence-based validation plan for AI Landing Zone Copilot assets, Bicep, parameters, scripts, and Azure deployment behavior. Use after implementation or during regression investigation."
}

AI Landing Zone validation

Map each acceptance criterion to an observable result.
Identify changed declarative assets, Bicep contracts, scripts, and deployment paths.
Run Copilot asset validation and its fixture tests when .github engineering assets change.
Run targeted PowerShell tests for deterministic script logic.
Build and lint Bicep; run the size gate when compiled output can change.
Run preflight with the appropriate deterministic or Azure-aware scope.
Use What-If and approved test deployment evidence according to risk.
Report exact commands, results, warnings, skipped paths, and residual risk.

Do not treat successful YAML parsing, Bicep compilation, or What-If as proof of runtime success. Do not mutate Azure merely to gather evidence without approval.