Phase 4: orchestrator + cycles auto-execute

Componente runtime/ che cabla core+clients+state+safety in un engine
autonomo notify-only: nessuna conferma manuale, ordini combo
piazzati direttamente quando le regole passano. 311 test pass,
copertura totale 94%, runtime/ 90%, mypy strict pulito, ruff clean.

Moduli:
- runtime/alert_manager.py: escalation tree
  LOW/MEDIUM/HIGH/CRITICAL → audit + Telegram + kill switch.
- runtime/dependencies.py: build_runtime() costruisce
  RuntimeContext con tutti i client MCP, repository, audit log,
  kill switch, alert manager.
- runtime/entry_cycle.py: flusso settimanale (snapshot parallelo
  spot/dvol/funding/macro/holdings/equity → validate_entry →
  compute_bias → options_chain → select_strikes →
  liquidity_gate → sizing_engine → combo_builder.build →
  place_combo_order → notify_position_opened).
- runtime/monitor_cycle.py: loop 12h con dvol_history per il
  return_4h, exit_decision.evaluate, close auto-execute.
- runtime/health_check.py: probe parallelo MCP + SQLite +
  environment match; 3 strikes consecutivi → kill switch HIGH.
- runtime/recovery.py: riconciliazione SQLite vs broker
  all'avvio; mismatch → kill switch CRITICAL.
- runtime/scheduler.py: AsyncIOScheduler builder con cron entry
  (lun 14:00), monitor (02/14), health (5min).
- runtime/orchestrator.py: façade boot() + run_entry/monitor/health
  + install_scheduler + run_forever, con env check vs strategy.

CLI:
- start: avvia engine bloccante (asyncio.run + scheduler).
- dry-run --cycle entry|monitor|health: esegue un singolo ciclo
  per debug/test in produzione.
- stop: documenta lo shutdown via SIGTERM al container.

Documentazione:
- docs/06-operational-flow.md riscritto per il modello
  notify-only auto-execute (no conferma manuale, no memory,
  no brain-bridge).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

tests/unit/test_runtime_dependencies.py

@@ -0,0 +1,55 @@
+"""Tests for the runtime dependency container."""
+
+from __future__ import annotations
+
+from datetime import UTC, datetime
+from pathlib import Path
+
+from cerbero_bite.config import golden_config
+from cerbero_bite.config.mcp_endpoints import load_endpoints
+from cerbero_bite.runtime import build_runtime
+from cerbero_bite.state import connect
+
+
+def test_build_runtime_creates_state_and_audit_files(tmp_path: Path) -> None:
+    db_path = tmp_path / "state.sqlite"
+    audit_path = tmp_path / "audit.log"
+
+    ctx = build_runtime(
+        cfg=golden_config(),
+        endpoints=load_endpoints(env={}),
+        token="t",
+        db_path=db_path,
+        audit_path=audit_path,
+        clock=lambda: datetime(2026, 4, 27, 14, 0, tzinfo=UTC),
+    )
+
+    assert db_path.exists()
+    assert ctx.audit_log.path == audit_path
+    # system_state singleton initialised
+    conn = connect(db_path)
+    try:
+        state = ctx.repository.get_system_state(conn)
+    finally:
+        conn.close()
+    assert state is not None
+    assert state.config_version == ctx.cfg.config_version
+
+
+def test_build_runtime_clients_pinned_to_endpoints(tmp_path: Path) -> None:
+    ctx = build_runtime(
+        cfg=golden_config(),
+        endpoints=load_endpoints(
+            env={"CERBERO_BITE_MCP_DERIBIT_URL": "http://localhost:9911"}
+        ),
+        token="t",
+        db_path=tmp_path / "state.sqlite",
+        audit_path=tmp_path / "audit.log",
+    )
+    # type checks: every client is the right concrete type
+    assert ctx.deribit.SERVICE == "deribit"
+    assert ctx.macro.SERVICE == "macro"
+    assert ctx.sentiment.SERVICE == "sentiment"
+    assert ctx.hyperliquid.SERVICE == "hyperliquid"
+    assert ctx.portfolio.SERVICE == "portfolio"
+    assert ctx.telegram.SERVICE == "telegram"