메뉴
Synthesize an Azure Monitor impact report for a Chaos Studio v2 ScenarioRun: pulls the run, walks targeted resources, queries metrics/logs/activity-log/alerts over the run window ± buffer, and renders a Markdown + JSON impact card.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/microsoft/chaos-studio --skill chaos-impact명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
Execute a ScenarioConfiguration and stream ScenarioRun status with per-action breakdown until terminal state.
Discover recommended scenarios, build and validate a ScenarioConfiguration, and auto-fix resource permissions.
Provision a Microsoft.Chaos/workspaces resource (v2), bind a managed identity, set scopes, and grant Reader RBAC on the scope.
SKILL.md 표시 중
SOC 직업 분류 기준
| name | chaos-impact |
| description | Synthesize an Azure Monitor impact report for a Chaos Studio v2 ScenarioRun: pulls the run, walks targeted resources, queries metrics/logs/activity-log/alerts over the run window ± buffer, and renders a Markdown + JSON impact card. |
⛔ ABSOLUTE RULE: Do NOT improvise, skip, or substitute any step. On ANY error, STOP and wait for the user.
This skill is the human-interactive path: it resolves run context from the
shared startchaos-state.json, prompts for missing inputs via exit codes, and
renders a Markdown card plus a JSON sidecar. Use it when there is a user in the
loop — typically right after /start-chaos (or as a post-step of it).
If you are an autonomous agent with no user to prompt, call the three
Monitor tools on the chaos-studio MCP server directly (monitor_query_metrics,
monitor_query_logs, monitor_search_activity_log) and assemble your own
report. The PowerShell skill and the MCP tools are two independent surfaces
over the same Azure endpoints — neither calls the other.
Both surfaces target Microsoft.Chaos 2026-05-01-preview plus the pinned
Azure Monitor API versions in scripts/Constants.ps1, and use the local
az login session for auth.
All synthesis logic lives in scripts/Invoke-ChaosImpact.ps1. The script
resolves the ScenarioRun, discovers Log Analytics workspaces via diagnostic
settings, runs the Monitor query fan-out, classifies signals, and writes both
artifacts.
The AI orchestrator's only job is:
$env:STARTCHAOS_STATE_PATH to
(the skill but does not this file).${SESSION_DIR}/startchaos-state.json-ScenarioRunId <id> (positional accepted).$env:STARTCHAOS_STATE_PATH = "<session-dir>/startchaos-state.json"
& "<skill-dir>/scripts/Invoke-ChaosImpact.ps1" -ScenarioRunId <runId> @extraArgs
When invoked standalone (no prior start-chaos state file), the orchestrator
must supply -SubscriptionId, -ResourceGroup, -WorkspaceName, and
-ScenarioName — otherwise the script exits 2.
| Exit | Meaning | AI Action |
|---|---|---|
| 0 | Report generated | Done — print the Markdown card from script output. |
| 1 | Hard error | STOP. Render the error from script output. Wait for user. |
| 2 | Missing run context | ask_user for any of: -SubscriptionId, -ResourceGroup, -WorkspaceName, -ScenarioName, -ScenarioRunId. Re-run with the supplied parameters. |
| 3 | Log Analytics workspace not discoverable for ≥ 1 resource, AND -LogAnalyticsWorkspaceId not supplied | ask_user whether to (a) supply a workspace ID (re-run with -LogAnalyticsWorkspaceId <id>) or (b) proceed without log signals (re-run with -LogAnalyticsWorkspaceId none). |
| 4 | Permission gap (caller lacks Monitoring Reader on one or more target resource groups) (Epic 3 — not yet implemented) | ask_user to confirm best-effort partial report (re-run with -AllowPartial) or abort. |
| Parameter | Description |
|---|---|
-ScenarioRunId | Required (positional). The ScenarioRun resource name. |
-SubscriptionId / -ResourceGroup / -WorkspaceName / -ScenarioName | Run context. Read from startchaos-state.json when omitted; required otherwise. |
-Buffer | ISO-8601 duration added before/after the run window. Default PT5M. |
-LogAnalyticsWorkspaceId | Optional override when diag-settings discovery fails for some resources. Pass none to explicitly skip log signals. |
-MaxResources | Per-run resource cap. Default 50. Instance-level targets (VMSS / AKS) are collapsed to parent before counting. |
-OutputDir | Where to drop artifacts. Default $env:STARTCHAOS_SESSION_DIR or current directory. |
-IncludeBaseline | When set, also computes baseline samples (the -Buffer × 2 window before run start). Default on. |
-Format | markdown, json, or both. Default both. |
-AllowPartial | Continue past permission gaps and emit a partial report. (Epic 3 — not yet implemented; currently accepted but has no effect.) |
MaxResources cap with explicit coverage caveats