run-scenario

# RunScenario — Scenario Execution & Status Streaming > ⛔ **ABSOLUTE RULE**: Do NOT improvise, skip, or substitute any step. On ANY error, STOP and wait for the user. ## When to use this skill (vs. the MCP server) 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. ## How It Works 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: 1. Set `$env:STARTCHAOS_STATE_PATH` to `${SESSION_DIR}/startchaos-state.json`. 2. Run the script. 3. Handle exit codes (see below). The script takes **no parameters**: configuration name, scenario, and workspace are all read from state. ## Prerequisites - `state.setup.status == "done"` (configuration must exist and be validated) - `state.setup.configuration.id` populated ## Running the Script ```powershell $env:STARTCHAOS_STATE_PATH = "<session-dir>/startchaos-state.json" & "<skill-dir>/scripts/Invoke-RunScenario.ps1" ``` For non-interactive sessions, set `$env:STARTCHAOS_NONINTERACTIVE=1` to skip the confirmation prompt. ## Exit Codes → AI Actions | 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`. | ## Cancellation On `Ctrl+C`, the script invokes `ScenarioRuns_Cancel` (POST `.../runs/{runId}/cancel`) best-effort and persists `state.run.lastError = "user-cancelled"`. ## What the Script Handles (no AI logic needed) - Confirmation card with scenario, parameters, and scope summary (suppressed when `STARTCHAOS_NONINTERACTIVE=1`) - Run start via `az chaos scenario run start --skip-validation --no-wait` (validation already gated upstream) - ScenarioRun ID resolution from the start result, with `az chaos scenario run list` fallback - Per-poll status via `az chaos scenario run show`: top status, elapsed time, per-action `scenarioRunSummary[]` table, resource count, error counts - Terminal-state detection (`Succeeded`, `Failed`, `Canceled`) and final summary card - Atomic state writes with error envelopes - Cancellation handling ## Related Skills - `setup-scenario` — must complete before this skill - `chaos-impact` — post-run Azure Monitor impact synthesis - `start-chaos` — orchestrator that invokes this skill