메뉴
Execute a ScenarioConfiguration and stream ScenarioRun status with per-action breakdown until terminal state.
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 run-scenario명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
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.
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.
SKILL.md 표시 중
SOC 직업 분류 기준
| name | run-scenario |
| description | Execute a ScenarioConfiguration and stream ScenarioRun status with per-action breakdown until terminal state. |
⛔ 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 reads run context from startchaos-state.json, renders status cards, and persists the run result. Use it when there is a user in the loop.
If you are an autonomous agent with no user to prompt, use chaos_execute_scenario + chaos_get_scenario_run (MCP) for a non-interactive run-and-poll flow. Cancellation: chaos_cancel_scenario_run. See mcp/README.md.
Both surfaces target Microsoft.Chaos 2026-05-01-preview and use the local az login session for auth.
All execution and status-streaming logic lives in scripts/Invoke-RunScenario.ps1. The script handles confirmation gating, starting the run via az chaos scenario run start --skip-validation --no-wait, ScenarioRun ID resolution (start result + run list fallback), per-poll status rendering via az chaos scenario run show, terminal-state detection, and the final summary.
The AI orchestrator's only job is:
$env:STARTCHAOS_STATE_PATH to ${SESSION_DIR}/startchaos-state.json.The script takes no parameters: configuration name, scenario, and workspace are all read from state.
state.setup.status == "done" (configuration must exist and be validated)state.setup.configuration.id populated$env:STARTCHAOS_STATE_PATH = "<session-dir>/startchaos-state.json"
& "<skill-dir>/scripts/Invoke-RunScenario.ps1"
For non-interactive sessions, set $env:STARTCHAOS_NONINTERACTIVE=1 to skip the confirmation prompt.
| Exit | Meaning | AI Action |
|---|---|---|
| 0 | Run reached a terminal state (Succeeded / Failed / Canceled) | Done — final card already rendered by script. Inspect state.run.status for terminal classification. |
| 1 | Hard error before/during execution | STOP. Render the error from script output. Wait for user. State has run.lastError. |
On Ctrl+C, the script invokes ScenarioRuns_Cancel (POST .../runs/{runId}/cancel) best-effort and persists state.run.lastError = "user-cancelled".
STARTCHAOS_NONINTERACTIVE=1)az chaos scenario run start --skip-validation --no-wait (validation already gated upstream)az chaos scenario run list fallbackaz chaos scenario run show: top status, elapsed time, per-action scenarioRunSummary[] table, resource count, error countsSucceeded, Failed, Canceled) and final summary cardsetup-scenario — must complete before this skillchaos-impact — post-run Azure Monitor impact synthesisstart-chaos — orchestrator that invokes this skill