Initial commit: SWOOSH bootstrap with statechart spec, OpenAPI, ADRs, and ops harnesses

ops/chaos-harness.md

@@ -0,0 +1,14 @@
+# Chaos Harness
+
+## Scenarios
+- Node churn (random kill/restart orchestrator replicas).
+- Message duplication and reordering on HMMM.
+- License server throttling (grace + degrade paths).
+- Index swap mid-formation (pin via content hash).
+
+## Metrics to Watch
+- state.advance_latency{region}
+- retry_count{transition}
+- quarantine_rate
+- recovery_time_from_churn
+- pending_transitions

ops/observability.md

@@ -0,0 +1,5 @@
+# Observability
+
+- Export Prometheus metrics for transitions, guards, retries.
+- Trace: one span per transition; include guard outcomes and linked evidence.
+- Logs: append-only structured logs with state_pre_hash and state_post_hash.