Tune EUR/USD defaults: tighter SL, higher TP, A-grade filter, unblock ranging

Config changes from EUR/USD optimization rounds:
- sl_atr_multiplier 1.5→0.5, tp_atr_multiplier 1.0→1.5, min_rr 0.5→1.0
- Unblock ranging regime, restrict to A+/A quality on M15
- Lower entry thresholds: score 38→30, confidence 60→50
- Add M5-specific regime weights and config overrides in eurusd.py

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

config.py

@@ -169,22 +169,22 @@ class EurUsdStrategyConfig:
     weekend_filter: bool = True
     allowed_weekdays: List[int] = field(default_factory=lambda: [0, 1, 2, 3, 4])
     min_atr_threshold: float = 0.0003
-    sl_atr_multiplier: float = 1.5
+    sl_atr_multiplier: float = 0.5
-    tp_atr_multiplier: float = 1.0
+    tp_atr_multiplier: float = 1.5
 
     sl_atr_floor: float = 1.2
-    min_rr_ratio: float = 0.5
+    min_rr_ratio: float = 1.0
     mature_move_threshold: float = 35.0
     tp_quality_scale: Dict[str, float] = field(default_factory=lambda: {
         "A+": 1.0, "A": 1.0, "B": 1.0, "C": 1.0,
     })
-    blocked_regimes: List[str] = field(default_factory=lambda: ["ranging"])
+    blocked_regimes: List[str] = field(default_factory=list)
-    min_quality_grades: List[str] = field(default_factory=lambda: ["A+", "A", "B"])
+    min_quality_grades: List[str] = field(default_factory=lambda: ["A+", "A"])
     max_exhaustion_tags: int = 99
     block_near_ema200: bool = False
     session_hours: List[int] = field(default_factory=list)
-    min_entry_score: float = 38.0
+    min_entry_score: float = 30.0
-    min_entry_confidence: float = 60.0
+    min_entry_confidence: float = 50.0
     block_rsi_extended: bool = True
     be_trigger_r: float = 0.3
     be_buffer_r: float = 0.05

engine.py

@@ -488,7 +488,7 @@ class TechnicalEngine:
         plus_di = 100 * (plus_dm.ewm(alpha=1/c.adx_period, min_periods=c.adx_period, adjust=False).mean() / atr)
         minus_di = 100 * (minus_dm.ewm(alpha=1/c.adx_period, min_periods=c.adx_period, adjust=False).mean() / atr)
 
-        dx = 100 * (plus_di - minus_di).abs() / (plus_di + minus_di).replace(0, np.nan)
+        dx = 100 * (plus_di - minus_di).abs() / np.maximum(plus_di + minus_di, 1e-10)
         adx = dx.ewm(alpha=1/c.adx_period, min_periods=c.adx_period, adjust=False).mean()
 
         adx_val = adx.iloc[-1] if not adx.empty and not pd.isna(adx.iloc[-1]) else 0.0
@@ -598,8 +598,12 @@ class TechnicalEngine:
         if len(close) < n + 5:
             return
 
-        roc_val = ((close.iloc[-1] - close.iloc[-n - 1]) / close.iloc[-n - 1]) * 100
+        denom1 = close.iloc[-n - 1]
-        roc_prev = ((close.iloc[-2] - close.iloc[-n - 2]) / close.iloc[-n - 2]) * 100
+        denom2 = close.iloc[-n - 2]
+        if denom1 == 0 or denom2 == 0:
+            return
+        roc_val = ((close.iloc[-1] - denom1) / denom1) * 100
+        roc_prev = ((close.iloc[-2] - denom2) / denom2) * 100
 
         if pd.isna(roc_val) or pd.isna(roc_prev):
             return
@@ -1112,12 +1116,14 @@ class TechnicalEngine:
         for i in range(window, len(vals) - window):
             if np.isnan(vals[i]):
                 continue
+            left_vals = [vals[i-j] for j in range(1, window+1) if not np.isnan(vals[i-j])]
+            right_vals = [vals[i+j] for j in range(1, window+1) if i+j < len(vals) and not np.isnan(vals[i+j])]
+            if not left_vals or not right_vals:
+                continue
             if mode == "low":
-                if all(vals[i] <= vals[i-j] for j in range(1, window+1) if not np.isnan(vals[i-j])) and \
+                if all(vals[i] <= v for v in left_vals) and all(vals[i] <= v for v in right_vals):
-                   all(vals[i] <= vals[i+j] for j in range(1, window+1) if i+j < len(vals) and not np.isnan(vals[i+j])):
                     points.append((i, vals[i]))
             else:
-                if all(vals[i] >= vals[i-j] for j in range(1, window+1) if not np.isnan(vals[i-j])) and \
+                if all(vals[i] >= v for v in left_vals) and all(vals[i] >= v for v in right_vals):
-                   all(vals[i] >= vals[i+j] for j in range(1, window+1) if i+j < len(vals) and not np.isnan(vals[i+j])):
                     points.append((i, vals[i]))
         return points

news_engine.py

@@ -497,11 +497,12 @@ class NewsEngine:
         found = []
 
         for currency, keywords in self.CURRENCY_KEYWORDS.items():
-            if currency.lower() in headline_lower:
+            if re.search(r'\b' + re.escape(currency.lower()) + r'\b', headline_lower):
                 found.append(currency)
                 continue
             for kw in keywords:
-                if kw in headline_lower:
+                kw_stripped = kw.strip()
+                if kw_stripped and re.search(r'\b' + re.escape(kw_stripped) + r'\b', headline_lower):
                     found.append(currency)
                     break
 

pair_signals/eurusd.py

@@ -32,6 +32,28 @@ EUR_REGIME_WEIGHTS = {
     },
 }
 
+_M5_TRENDING = {
+    "ema_trend": 1.5, "rsi": 1.0, "macd": 2.5,
+    "bollinger": 0.5, "keltner": 2.0, "roc": 2.0,
+    "candlestick": 0.3, "divergence": 1.0, "adx": 1.0,
+}
+
+M5_REGIME_WEIGHTS = {
+    "trending_up": _M5_TRENDING,
+    "trending_down": _M5_TRENDING,
+    "trending": _M5_TRENDING,
+    "ranging": {
+        "ema_trend": 0.3, "rsi": 1.5, "macd": 1.5,
+        "bollinger": 1.5, "keltner": 2.0, "roc": 1.0,
+        "candlestick": 0.3, "divergence": 1.5, "adx": 0.5,
+    },
+    "transitioning": {
+        "ema_trend": 1.0, "rsi": 1.0, "macd": 2.5,
+        "bollinger": 0.5, "keltner": 2.5, "roc": 2.0,
+        "candlestick": 0.3, "divergence": 1.0, "adx": 1.0,
+    },
+}
+
 
 def make_config(
     primary_tf: Timeframe,
@@ -53,6 +75,14 @@ def make_config(
     config.backtest.blocked_hours_utc = [22, 23, 0, 1, 2, 3]
     config.backtest.allowed_weekdays = [0, 1, 2, 3, 4]
 
-    config.technical.regime_weights_override = EUR_REGIME_WEIGHTS
+    if primary_tf == Timeframe.M5:
+        config.technical.regime_weights_override = M5_REGIME_WEIGHTS
+        config.eur_strategy.sl_atr_floor = 0.8
+        config.eur_strategy.sl_atr_multiplier = 0.3
+        config.eur_strategy.min_quality_grades = ["A+", "A", "B"]
+        config.eur_strategy.min_entry_score = 28.0
+        config.eur_strategy.min_entry_confidence = 50.0
+    else:
+        config.technical.regime_weights_override = EUR_REGIME_WEIGHTS
 
     return config

signal_aggregator.py

@@ -252,11 +252,11 @@ class SignalAggregator:
             )
             report.trade_setup = None
         else:
-            report.bullish_reasons.append(
+            reason = f"Regime: {regime_tf.value} ADX={adx_value:.1f} confirms trending market"
-                f"Regime: {regime_tf.value} ADX={adx_value:.1f} confirms trending market"
+            if report.final_score > 0:
-            ) if report.final_score > 0 else report.bearish_reasons.append(
+                report.bullish_reasons.append(reason)
-                f"Regime: {regime_tf.value} ADX={adx_value:.1f} confirms trending market"
+            else:
-            )
+                report.bearish_reasons.append(reason)
 
     @staticmethod
     def _latest_adx(df: pd.DataFrame, period: int = 14) -> Optional[float]:
@@ -915,6 +915,8 @@ class SignalAggregator:
             )
             return None
 
+        if pip_value <= 0:
+            return None
         min_sl_pips = 8.0 if pip_value < 0.01 else 80.0
         min_sl_price = min_sl_pips * pip_value
 
@@ -932,9 +934,11 @@ class SignalAggregator:
             eur = self.config.eur_strategy
             min_rr = eur.min_rr_ratio
             mature_move = abs(report.final_score) > eur.mature_move_threshold
-            tp_mult = eur.tp_atr_multiplier if mature_move else eur.tp_atr_multiplier * 1.5
+            quality_scale = eur.tp_quality_scale.get(report.signal_quality, 1.0)
-            tp_swing_cap = 2.0 if mature_move else 2.5
+            tp_mult = (eur.tp_atr_multiplier if mature_move else eur.tp_atr_multiplier * 1.5) * quality_scale
+            tp_swing_cap = (2.0 if mature_move else 2.5) * quality_scale
             sl_atr_floor = eur.sl_atr_floor
+            sl_swing_buffer = eur.sl_atr_multiplier
         else:
             min_rr = self.config.signals.min_rr_ratio
             mature_move = abs(report.final_score) > 30
@@ -942,6 +946,7 @@ class SignalAggregator:
             tp_swing_cap = 2.0 if mature_move else 2.5
             sl_atr_floor = (self.config.xau_strategy.v2_sl_atr_multiplier
                             if report.display_name == "XAU/USD" else 1.5)
+            sl_swing_buffer = 1.0
 
         if is_buy:
             swing_lows = sorted(
@@ -949,9 +954,9 @@ class SignalAggregator:
                 reverse=True,
             )
             if swing_lows:
-                sl_price = swing_lows[0] - atr * 1.0
+                sl_price = swing_lows[0] - atr * sl_swing_buffer
             else:
-                sl_price = price - atr * 2.0
+                sl_price = price - atr * (1.0 + sl_swing_buffer)
             min_sl = max(atr * sl_atr_floor, min_sl_price)
             if price - sl_price < min_sl:
                 sl_price = price - min_sl
@@ -979,9 +984,9 @@ class SignalAggregator:
                 v for k, v in levels.items() if k.startswith("swing_high_") and v > price
             )
             if swing_highs:
-                sl_price = swing_highs[0] + atr * 1.0
+                sl_price = swing_highs[0] + atr * sl_swing_buffer
             else:
-                sl_price = price + atr * 2.0
+                sl_price = price + atr * (1.0 + sl_swing_buffer)
             min_sl = max(atr * sl_atr_floor, min_sl_price)
             if sl_price - price < min_sl:
                 sl_price = price + min_sl

test_all.py

@@ -1677,14 +1677,15 @@ def _():
     assert 6 not in cfg.allowed_weekdays  # No Sunday
     assert cfg.min_atr_threshold == 0.0003
     assert cfg.sl_atr_floor == 1.2
-    assert cfg.min_rr_ratio == 0.5
+    assert cfg.min_rr_ratio == 1.0
     assert cfg.mature_move_threshold == 35.0
     assert cfg.tp_quality_scale["A+"] == 1.0
     assert cfg.tp_quality_scale["A"] == 1.0
     assert cfg.tp_quality_scale["C"] == 1.0
-    assert "ranging" in cfg.blocked_regimes
+    assert cfg.blocked_regimes == []
     assert "A" in cfg.min_quality_grades
-    assert "C" not in cfg.min_quality_grades
+    assert "A" in cfg.min_quality_grades
+    assert "B" not in cfg.min_quality_grades
 
 
 @test("PlatformConfig has eur_strategy field")
@@ -1776,7 +1777,7 @@ def _():
     config.eur_strategy.blocked_regimes = []
     config.eur_strategy.min_quality_grades = []
     assert config.eur_strategy.sl_atr_floor == 1.2
-    assert config.eur_strategy.min_rr_ratio == 0.5
+    assert config.eur_strategy.min_rr_ratio == 1.0
     agg = SignalAggregator(config)
     df = make_ohlcv(n=300, start_price=1.0800, trend=0.0003, seed=77)
     dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
@@ -1785,6 +1786,971 @@ def _():
         assert report.trade_setup.risk_reward >= 0.5 - 0.01
 
 
+# ══════════════════════════════════════════════════════════════════════
+#  MODULE 10: Bug Fix Regressions & Comprehensive Edge Cases
+# ══════════════════════════════════════════════════════════════════════
+
+print("\n═══ Module 10: Bug Fix Regressions & Edge Cases ═══")
+
+# ── 10a. Swing Point Bug Fix (engine.py) ─────────────────────────────
+
+@test("Swing points: all-NaN neighbors produces no false positives")
+def _():
+    series = pd.Series([np.nan, np.nan, np.nan, 5.0, np.nan, np.nan, np.nan])
+    lows = TechnicalEngine._find_swing_points(series, mode="low", window=3)
+    highs = TechnicalEngine._find_swing_points(series, mode="high", window=3)
+    assert lows == [], f"Expected no lows with all-NaN neighbors, got {lows}"
+    assert highs == [], f"Expected no highs with all-NaN neighbors, got {highs}"
+
+
+@test("Swing points: partial NaN neighbors still detect valid swings")
+def _():
+    series = pd.Series([10.0, 8.0, np.nan, 4.0, np.nan, 8.0, 10.0])
+    lows = TechnicalEngine._find_swing_points(series, mode="low", window=2)
+    low_vals = [v for _, v in lows]
+    assert 4.0 in low_vals, f"Expected 4.0 as swing low, got {low_vals}"
+
+
+@test("Swing points: empty series returns empty list")
+def _():
+    series = pd.Series([], dtype=float)
+    assert TechnicalEngine._find_swing_points(series, mode="low") == []
+    assert TechnicalEngine._find_swing_points(series, mode="high") == []
+
+
+@test("Swing points: series shorter than 2*window returns empty")
+def _():
+    series = pd.Series([1.0, 2.0, 3.0])
+    assert TechnicalEngine._find_swing_points(series, mode="low", window=3) == []
+
+
+@test("Swing points: monotonic increasing has no interior lows")
+def _():
+    series = pd.Series([float(i) for i in range(20)])
+    lows = TechnicalEngine._find_swing_points(series, mode="low", window=2)
+    assert len(lows) == 0, f"Monotonic increase should have no interior lows, got {lows}"
+
+
+@test("Swing points: monotonic decreasing has no interior highs")
+def _():
+    series = pd.Series([float(20 - i) for i in range(20)])
+    highs = TechnicalEngine._find_swing_points(series, mode="high", window=2)
+    assert len(highs) == 0, f"Monotonic decrease should have no interior highs, got {highs}"
+
+
+@test("Swing points: V-shape detects low at bottom")
+def _():
+    series = pd.Series([10.0, 8.0, 6.0, 4.0, 6.0, 8.0, 10.0])
+    lows = TechnicalEngine._find_swing_points(series, mode="low", window=2)
+    low_vals = [v for _, v in lows]
+    assert 4.0 in low_vals, f"V-shape should detect 4.0 as low, got {low_vals}"
+
+
+@test("Swing points: inverted V-shape detects high at top")
+def _():
+    series = pd.Series([4.0, 6.0, 8.0, 10.0, 8.0, 6.0, 4.0])
+    highs = TechnicalEngine._find_swing_points(series, mode="high", window=2)
+    high_vals = [v for _, v in highs]
+    assert 10.0 in high_vals, f"Inverted V should detect 10.0 as high, got {high_vals}"
+
+
+# ── 10b. ROC Bug Fix (engine.py) ─────────────────────────────────────
+
+@test("ROC: zero close at lookback position does not produce inf")
+def _():
+    df = make_ohlcv(n=100, seed=42)
+    df.iloc[-11, df.columns.get_loc("Close")] = 0.0
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    if "roc" in result.indicators:
+        assert np.isfinite(result.indicators["roc"].value), "ROC should not be inf"
+
+
+@test("ROC: normal computation produces finite values")
+def _():
+    df = make_ohlcv(n=100, seed=55)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    if "roc" in result.indicators:
+        assert np.isfinite(result.indicators["roc"].value)
+        assert 0 <= result.indicators["roc"].confidence <= 100
+
+
+# ── 10c. ADX Epsilon Fix (engine.py) ────────────────────────────────
+
+@test("ADX: flat data produces finite ADX value")
+def _():
+    n = 100
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    df = pd.DataFrame({
+        "Open": [100.0]*n, "High": [100.01]*n, "Low": [99.99]*n,
+        "Close": [100.0]*n, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    if "adx" in result.indicators:
+        assert np.isfinite(result.indicators["adx"].value), "ADX should be finite on flat data"
+
+
+@test("ADX: strong uptrend shows high ADX")
+def _():
+    df = make_ohlcv(n=300, trend=0.001, seed=10)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert "adx" in result.indicators
+    assert result.indicators["adx"].value > 0
+
+
+# ── 10d. RSI Boundaries ─────────────────────────────────────────────
+
+@test("RSI: monotonic up approaches 100")
+def _():
+    n = 100
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    close = np.linspace(100, 150, n)
+    df = pd.DataFrame({
+        "Open": close - 0.1, "High": close + 0.5, "Low": close - 0.5,
+        "Close": close, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert "rsi" in result.indicators
+    assert result.indicators["rsi"].value > 80, f"RSI should be high on uptrend, got {result.indicators['rsi'].value}"
+
+
+@test("RSI: monotonic down approaches 0")
+def _():
+    n = 100
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    close = np.linspace(150, 100, n)
+    df = pd.DataFrame({
+        "Open": close + 0.1, "High": close + 0.5, "Low": close - 0.5,
+        "Close": close, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert "rsi" in result.indicators
+    assert result.indicators["rsi"].value < 20, f"RSI should be low on downtrend, got {result.indicators['rsi'].value}"
+
+
+@test("RSI: flat price produces RSI near 50")
+def _():
+    n = 100
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    np.random.seed(99)
+    close = 100.0 + np.random.randn(n) * 0.001
+    df = pd.DataFrame({
+        "Open": close, "High": close + 0.01, "Low": close - 0.01,
+        "Close": close, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    if "rsi" in result.indicators:
+        rsi = result.indicators["rsi"].value
+        assert 30 <= rsi <= 70, f"RSI on flat should be near 50, got {rsi}"
+
+
+# ── 10e. MACD Edge Cases ────────────────────────────────────────────
+
+@test("MACD: very short data (< 35 bars) skips cleanly")
+def _():
+    df = make_ohlcv(n=30, seed=42)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    # Should not crash; macd may or may not be present
+
+
+@test("MACD: strong uptrend produces non-SELL signal")
+def _():
+    df = make_ohlcv(n=300, trend=0.001, seed=20)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    if "macd" in result.indicators:
+        assert result.indicators["macd"].signal not in (SignalStrength.SELL, SignalStrength.STRONG_SELL), \
+            f"MACD on uptrend should not be bearish, got {result.indicators['macd'].signal}"
+
+
+# ── 10f. Bollinger Bands Edge Cases ─────────────────────────────────
+
+@test("Bollinger: flat price collapses bands, position near 0.5")
+def _():
+    n = 100
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    df = pd.DataFrame({
+        "Open": [100.0]*n, "High": [100.01]*n, "Low": [99.99]*n,
+        "Close": [100.0]*n, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    if "bollinger" in result.indicators:
+        pos = result.indicators["bollinger"].value
+        assert 0.0 <= pos <= 1.0, f"BB position should be in [0,1], got {pos}"
+
+
+# ── 10g. Keltner Channel Edge Cases ────────────────────────────────
+
+@test("Keltner: BB disabled does not crash squeeze detection")
+def _():
+    df = make_ohlcv(n=200, seed=42)
+    config = PlatformConfig()
+    config.technical.disabled_indicators = ["bollinger"]
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert "keltner" in result.indicators, "Keltner should still work with BB disabled"
+
+
+# ── 10h. Candlestick Pattern Edge Cases ─────────────────────────────
+
+@test("Candlestick: zero-range bar (H==L) does not crash")
+def _():
+    df = make_ohlcv(n=100, seed=42)
+    df.iloc[-1, df.columns.get_loc("High")] = df.iloc[-1]["Close"]
+    df.iloc[-1, df.columns.get_loc("Low")] = df.iloc[-1]["Close"]
+    df.iloc[-1, df.columns.get_loc("Open")] = df.iloc[-1]["Close"]
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert result is not None
+
+
+@test("Candlestick: very short data (< 5 bars) does not crash")
+def _():
+    df = make_ohlcv(n=4, seed=42)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert result is not None
+
+
+# ── 10i. Regime Classification ──────────────────────────────────────
+
+@test("Regime: strong uptrend produces trending_up or trending")
+def _():
+    df = make_ohlcv(n=300, trend=0.002, seed=15)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert result.market_regime in ("trending_up", "trending", "transitioning"), \
+        f"Strong uptrend should be trending, got {result.market_regime}"
+
+
+@test("Regime: strong downtrend produces trending_down or trending")
+def _():
+    df = make_ohlcv(n=300, trend=-0.002, seed=16)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert result.market_regime in ("trending_down", "trending", "transitioning"), \
+        f"Strong downtrend should be trending, got {result.market_regime}"
+
+
+@test("Regime: flat/choppy data produces ranging or transitioning")
+def _():
+    n = 300
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    np.random.seed(17)
+    close = 100.0 + np.random.randn(n) * 0.001
+    df = pd.DataFrame({
+        "Open": close, "High": close + 0.002, "Low": close - 0.002,
+        "Close": close, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert result.market_regime in ("ranging", "transitioning"), \
+        f"Flat data should be ranging, got {result.market_regime}"
+
+
+# ── 10j. Composite Score ────────────────────────────────────────────
+
+@test("Composite score always clamped to [-100, 100]")
+def _():
+    for seed in [1, 2, 3, 10, 20, 30]:
+        df = make_ohlcv(n=300, trend=0.005, seed=seed)
+        config = PlatformConfig()
+        engine = TechnicalEngine(config.technical)
+        result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+        assert -100 <= result.composite_score <= 100, \
+            f"Score {result.composite_score} out of bounds (seed={seed})"
+
+
+@test("Composite: regime weights affect score magnitude")
+def _():
+    df = make_ohlcv(n=300, trend=0.001, seed=42)
+    config1 = PlatformConfig()
+    config2 = PlatformConfig()
+    config2.technical.regime_weights_override = {
+        "trending_up": {"ema_trend": 0.1, "rsi": 0.1, "macd": 0.1,
+                        "bollinger": 0.1, "keltner": 0.1, "roc": 0.1,
+                        "candlestick": 0.1, "divergence": 0.1, "adx": 0.1},
+        "trending_down": {"ema_trend": 0.1, "rsi": 0.1, "macd": 0.1,
+                          "bollinger": 0.1, "keltner": 0.1, "roc": 0.1,
+                          "candlestick": 0.1, "divergence": 0.1, "adx": 0.1},
+        "ranging": {"ema_trend": 0.1, "rsi": 0.1, "macd": 0.1,
+                    "bollinger": 0.1, "keltner": 0.1, "roc": 0.1,
+                    "candlestick": 0.1, "divergence": 0.1, "adx": 0.1},
+        "transitioning": {"ema_trend": 0.1, "rsi": 0.1, "macd": 0.1,
+                          "bollinger": 0.1, "keltner": 0.1, "roc": 0.1,
+                          "candlestick": 0.1, "divergence": 0.1, "adx": 0.1},
+    }
+    engine1 = TechnicalEngine(config1.technical)
+    engine2 = TechnicalEngine(config2.technical)
+    r1 = engine1.analyze(df, "EUR/USD", Timeframe.H1)
+    r2 = engine2.analyze(df, "EUR/USD", Timeframe.H1)
+    assert r1.composite_score != r2.composite_score or r1.market_regime != r2.market_regime, \
+        "Different regime weights should produce different results"
+
+
+# ── 10k. Divergence Edge Cases ──────────────────────────────────────
+
+@test("Divergence: short lookback finds no divergence")
+def _():
+    df = make_ohlcv(n=100, seed=42)
+    config = PlatformConfig()
+    config.technical.divergence_lookback = 5
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    # Should not crash, divergence may or may not appear
+
+
+@test("Divergence: disabled divergence not in indicators")
+def _():
+    df = make_ohlcv(n=300, seed=42)
+    config = PlatformConfig()
+    config.technical.disabled_indicators = ["divergence"]
+    engine = TechnicalEngine(config.technical)
+    result = engine.analyze(df, "EUR/USD", Timeframe.H1)
+    assert "divergence" not in result.indicators
+
+
+# ══════════════════════════════════════════════════════════════════════
+#  MODULE 11: News Engine Edge Cases
+# ══════════════════════════════════════════════════════════════════════
+
+print("\n═══ Module 11: News Engine Edge Cases ═══")
+
+from news_engine import NewsEngine
+
+@test("Currency extraction: USDA does not false-match USD")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    result = ne._extract_currencies("USDA releases crop report")
+    assert "USD" not in result, f"USDA should not match USD, got {result}"
+
+
+@test("Currency extraction: custodian does not false-match USD")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    result = ne._extract_currencies("custodian reports quarterly earnings")
+    assert "USD" not in result, f"custodian should not match USD, got {result}"
+
+
+@test("Currency extraction: euro matches EUR correctly")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    result = ne._extract_currencies("euro strengthens against yen")
+    assert "EUR" in result, f"euro should match EUR, got {result}"
+
+
+@test("Currency extraction: Eurozone matches EUR")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    result = ne._extract_currencies("Eurozone GDP beats forecast")
+    assert "EUR" in result, f"Eurozone should match EUR, got {result}"
+
+
+@test("Currency extraction: empty headline returns empty list")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    result = ne._extract_currencies("")
+    assert result == [], f"Empty headline should return empty list, got {result}"
+
+
+@test("Currency extraction: headline with multiple currencies")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    result = ne._extract_currencies("EUR/USD pair rises as dollar weakens")
+    assert "EUR" in result, f"Should match EUR, got {result}"
+    assert "USD" in result, f"Should match USD, got {result}"
+
+
+@test("Sentiment: both bullish and bearish keywords gives bounded score")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    score = ne._analyze_headline_sentiment("rate hike fears amid bullish recovery expectations")
+    assert -1.0 <= score <= 1.0, f"Score should be bounded, got {score}"
+
+
+@test("Sentiment: empty headline returns 0.0")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    assert ne._analyze_headline_sentiment("") == 0.0
+
+
+@test("Sentiment: very long headline does not crash")
+def _():
+    ne = NewsEngine(PlatformConfig().news)
+    long_headline = "market " * 500 + "bullish recovery"
+    score = ne._analyze_headline_sentiment(long_headline)
+    assert isinstance(score, float)
+
+
+# ══════════════════════════════════════════════════════════════════════
+#  MODULE 12: Signal Aggregator Edge Cases
+# ══════════════════════════════════════════════════════════════════════
+
+print("\n═══ Module 12: Signal Aggregator Edge Cases ═══")
+
+# ── 12a. Quality Grade Boundaries ────────────────────────────────────
+
+@test("Quality grade: A+ requires score>=40, confidence>=65, MTF agreement")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, trend=0.003, seed=100)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.signal_quality == "A+":
+        assert abs(report.final_score) >= 40
+        assert report.final_confidence >= 65
+        assert report.mtf_agreement is True
+
+
+@test("Quality grade: A requires score>=33, confidence>=55, MTF")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, trend=0.002, seed=200)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.signal_quality == "A":
+        assert abs(report.final_score) >= 33
+        assert report.final_confidence >= 55
+        assert report.mtf_agreement is True
+
+
+@test("Quality grade: B requires score>=28, confidence>=50")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=300)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.signal_quality == "B":
+        assert abs(report.final_score) >= 28
+        assert report.final_confidence >= 50
+
+
+@test("Quality grade: C requires score>=25, confidence>=45")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=400)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.signal_quality == "C":
+        assert abs(report.final_score) >= 25
+        assert report.final_confidence >= 45
+
+
+@test("Quality grade: D requires score>=15")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=500)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.signal_quality == "D":
+        assert abs(report.final_score) >= 15
+
+
+@test("Quality grade: F has score<15 or zero confidence")
+def _():
+    n = 300
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    np.random.seed(600)
+    close = 100.0 + np.random.randn(n) * 0.0001
+    df = pd.DataFrame({
+        "Open": close, "High": close + 0.001, "Low": close - 0.001,
+        "Close": close, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.signal_quality == "F":
+        assert abs(report.final_score) < 15 or report.final_confidence < 45
+
+
+@test("Quality grade: all grades belong to valid set")
+def _():
+    valid_grades = {"A+", "A", "B", "C", "D", "F"}
+    for seed in range(42, 62):
+        df = make_ohlcv(n=300, seed=seed, trend=0.001 * (seed % 5 - 2))
+        config = PlatformConfig()
+        agg = SignalAggregator(config)
+        dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+        report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+        assert report.signal_quality in valid_grades, \
+            f"Unexpected grade '{report.signal_quality}' (seed={seed})"
+
+
+# ── 12b. EUR Filter Pipeline ────────────────────────────────────────
+
+@test("EUR filter: blocked regime sets quality F")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    config = eur_make_config(Timeframe.H1)
+    config.eur_strategy.blocked_regimes = ["ranging"]
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=42)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=EURUSD, dfs=dfs, use_news=False)
+    if report.market_regime == "ranging":
+        assert report.signal_quality == "F", \
+            f"Ranging should be blocked, got {report.signal_quality}"
+
+
+@test("EUR filter: quality not in min_quality_grades sets F")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    config = eur_make_config(Timeframe.H1)
+    config.eur_strategy.min_quality_grades = ["A+"]
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=42, trend=0.001)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=EURUSD, dfs=dfs, use_news=False)
+    if report.signal_quality not in ("A+", "F"):
+        assert False, f"Non-A+ grade should be blocked, got {report.signal_quality}"
+
+
+@test("EUR filter: score below min sets F")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    config = eur_make_config(Timeframe.H1)
+    config.eur_strategy.min_entry_score = 99.0
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=42, trend=0.001)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=EURUSD, dfs=dfs, use_news=False)
+    assert report.signal_quality == "F", \
+        f"Score 99 threshold should block all, got {report.signal_quality}"
+
+
+@test("EUR filter: confidence below min sets F")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    config = eur_make_config(Timeframe.H1)
+    config.eur_strategy.min_entry_confidence = 99.0
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=42, trend=0.001)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=EURUSD, dfs=dfs, use_news=False)
+    assert report.signal_quality == "F", \
+        f"Confidence 99 threshold should block all, got {report.signal_quality}"
+
+
+@test("EUR filter: RSI extended buy (RSI>70) blocked")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    config = eur_make_config(Timeframe.H1)
+    config.eur_strategy.block_rsi_extended = True
+    config.eur_strategy.min_entry_score = 0.0
+    config.eur_strategy.min_entry_confidence = 0.0
+    config.eur_strategy.min_quality_grades = ["A+", "A", "B", "C", "D"]
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, trend=0.005, seed=42)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=EURUSD, dfs=dfs, use_news=False)
+    if report.primary_analysis and "rsi" in report.primary_analysis.indicators:
+        rsi_v = report.primary_analysis.indicators["rsi"].value
+        if report.final_score > 0 and rsi_v > 70:
+            assert report.signal_quality == "F", \
+                f"Buy with RSI {rsi_v:.0f} should be blocked"
+
+
+# ── 12c. Trade Setup Edge Cases ─────────────────────────────────────
+
+@test("Trade setup: ATR zero returns None")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, trend=0.002, seed=42)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.primary_analysis:
+        report.primary_analysis.key_levels["atr"] = 0.0
+        setup = agg._generate_trade_setup(report, pip_value=config.pip_value)
+        assert setup is None, "ATR=0 should produce no trade setup"
+
+
+@test("Trade setup: pip_value zero returns None")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, trend=0.002, seed=42)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if report.primary_analysis and report.final_signal != SignalStrength.NEUTRAL:
+        setup = agg._generate_trade_setup(report, pip_value=0.0)
+        assert setup is None, "pip_value=0 should produce no trade setup"
+
+
+@test("Trade setup: BUY SL always below entry")
+def _():
+    for seed in [10, 20, 30, 40, 50]:
+        df = make_ohlcv(n=300, trend=0.002, seed=seed)
+        config = PlatformConfig()
+        agg = SignalAggregator(config)
+        dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+        report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+        if report.trade_setup and report.trade_setup.direction == "BUY":
+            assert report.trade_setup.stop_loss < report.trade_setup.entry_price, \
+                f"BUY SL should be below entry (seed={seed})"
+
+
+@test("Trade setup: SELL SL always above entry")
+def _():
+    for seed in [10, 20, 30, 40, 50]:
+        df = make_ohlcv(n=300, trend=-0.002, seed=seed)
+        config = PlatformConfig()
+        agg = SignalAggregator(config)
+        dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+        report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+        if report.trade_setup and report.trade_setup.direction == "SELL":
+            assert report.trade_setup.stop_loss > report.trade_setup.entry_price, \
+                f"SELL SL should be above entry (seed={seed})"
+
+
+@test("Trade setup: risk_reward is always positive when present")
+def _():
+    for seed in range(10, 30):
+        df = make_ohlcv(n=300, trend=0.001 * (seed % 5 - 2), seed=seed)
+        config = PlatformConfig()
+        agg = SignalAggregator(config)
+        dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+        report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+        if report.trade_setup:
+            assert report.trade_setup.risk_reward > 0, \
+                f"RR should be positive, got {report.trade_setup.risk_reward} (seed={seed})"
+
+
+@test("Trade setup: EUR uses EUR-specific parameters")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    config = eur_make_config(Timeframe.H1)
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, trend=0.002, start_price=1.08, seed=77)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=EURUSD, dfs=dfs, use_news=False)
+    if report.trade_setup:
+        assert report.trade_setup.risk_reward >= config.eur_strategy.min_rr_ratio - 0.01
+
+
+# ── 12d. MTF Agreement ──────────────────────────────────────────────
+
+@test("MTF: all NEUTRAL returns False")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    report = SignalReport(symbol="XAU/USD", display_name="XAU/USD", timestamp="now")
+    report.mtf_signals = {
+        "15min": SignalStrength.NEUTRAL,
+        "1h": SignalStrength.NEUTRAL,
+        "4h": SignalStrength.NEUTRAL,
+    }
+    assert agg._check_mtf_agreement(report) is False
+
+
+@test("MTF: 2 bearish + 1 bullish returns True (bearish agreement)")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    report = SignalReport(symbol="XAU/USD", display_name="XAU/USD", timestamp="now")
+    report.mtf_signals = {
+        "15min": SignalStrength.SELL,
+        "1h": SignalStrength.STRONG_SELL,
+        "4h": SignalStrength.BUY,
+    }
+    assert agg._check_mtf_agreement(report) is True
+
+
+@test("MTF: all same direction returns True")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    report = SignalReport(symbol="XAU/USD", display_name="XAU/USD", timestamp="now")
+    report.mtf_signals = {
+        "15min": SignalStrength.BUY,
+        "1h": SignalStrength.BUY,
+        "4h": SignalStrength.STRONG_BUY,
+    }
+    assert agg._check_mtf_agreement(report) is True
+
+
+@test("MTF: empty signals returns False")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    report = SignalReport(symbol="XAU/USD", display_name="XAU/USD", timestamp="now")
+    report.mtf_signals = {}
+    assert agg._check_mtf_agreement(report) is False
+
+
+# ── 12e. Final Signal Computation ───────────────────────────────────
+
+@test("Final signal: STRONG_BUY requires score>=35, confidence>=55")
+def _():
+    for seed in range(42, 62):
+        df = make_ohlcv(n=300, trend=0.001 * (seed % 4), seed=seed)
+        config = PlatformConfig()
+        agg = SignalAggregator(config)
+        dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+        report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+        if report.final_signal == SignalStrength.STRONG_BUY:
+            assert report.final_score >= 35, f"STRONG_BUY needs score>=35, got {report.final_score}"
+            assert report.final_confidence >= 55, f"STRONG_BUY needs conf>=55, got {report.final_confidence}"
+
+
+@test("Final signal: STRONG_SELL requires score<=-35, confidence>=55")
+def _():
+    for seed in range(42, 62):
+        df = make_ohlcv(n=300, trend=-0.001 * (seed % 4), seed=seed)
+        config = PlatformConfig()
+        agg = SignalAggregator(config)
+        dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+        report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+        if report.final_signal == SignalStrength.STRONG_SELL:
+            assert report.final_score <= -35, f"STRONG_SELL needs score<=-35, got {report.final_score}"
+            assert report.final_confidence >= 55, f"STRONG_SELL needs conf>=55, got {report.final_confidence}"
+
+
+@test("Final signal: NEUTRAL when score is low regardless of confidence")
+def _():
+    n = 300
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    np.random.seed(42)
+    close = 100.0 + np.random.randn(n) * 0.0001
+    df = pd.DataFrame({
+        "Open": close, "High": close + 0.001, "Low": close - 0.001,
+        "Close": close, "Volume": [1000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    if abs(report.final_score) < 25:
+        assert report.final_signal == SignalStrength.NEUTRAL
+
+
+# ── 12f. Penalty System ─────────────────────────────────────────────
+
+@test("Penalty: confidence never exceeds 95 or goes below 0")
+def _():
+    for seed in range(42, 72):
+        df = make_ohlcv(n=300, trend=0.001 * (seed % 6 - 3), seed=seed)
+        config = PlatformConfig()
+        agg = SignalAggregator(config)
+        dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+        report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+        assert 0.0 <= report.final_confidence <= 95.0, \
+            f"Confidence out of [0,95] range: {report.final_confidence} (seed={seed})"
+
+
+# ── 12g. Weekend Filter ─────────────────────────────────────────────
+
+@test("Weekend filter: Friday 22:01 UTC is weekend")
+def _():
+    from signal_aggregator import _is_forex_weekend
+    dt = datetime(2026, 5, 15, 22, 1, tzinfo=timezone.utc)  # Friday
+    assert dt.weekday() == 4
+    assert _is_forex_weekend(dt) is True
+
+
+@test("Weekend filter: Friday 21:59 UTC is NOT weekend")
+def _():
+    from signal_aggregator import _is_forex_weekend
+    dt = datetime(2026, 5, 15, 21, 59, tzinfo=timezone.utc)  # Friday
+    assert dt.weekday() == 4
+    assert _is_forex_weekend(dt) is False
+
+
+@test("Weekend filter: Saturday noon is weekend")
+def _():
+    from signal_aggregator import _is_forex_weekend
+    dt = datetime(2026, 5, 16, 12, 0, tzinfo=timezone.utc)  # Saturday
+    assert dt.weekday() == 5
+    assert _is_forex_weekend(dt) is True
+
+
+@test("Weekend filter: Sunday 21:59 UTC is weekend")
+def _():
+    from signal_aggregator import _is_forex_weekend
+    dt = datetime(2026, 5, 17, 21, 59, tzinfo=timezone.utc)  # Sunday
+    assert dt.weekday() == 6
+    assert _is_forex_weekend(dt) is True
+
+
+@test("Weekend filter: Sunday 22:00 UTC is NOT weekend")
+def _():
+    from signal_aggregator import _is_forex_weekend
+    dt = datetime(2026, 5, 17, 22, 0, tzinfo=timezone.utc)  # Sunday
+    assert dt.weekday() == 6
+    assert _is_forex_weekend(dt) is False
+
+
+@test("Weekend filter: Monday is not weekend")
+def _():
+    from signal_aggregator import _is_forex_weekend
+    dt = datetime(2026, 5, 18, 10, 0, tzinfo=timezone.utc)  # Monday
+    assert dt.weekday() == 0
+    assert _is_forex_weekend(dt) is False
+
+
+# ══════════════════════════════════════════════════════════════════════
+#  MODULE 13: Integration Edge Cases
+# ══════════════════════════════════════════════════════════════════════
+
+print("\n═══ Module 13: Integration Edge Cases ═══")
+
+
+@test("Integration: all indicators disabled produces valid report")
+def _():
+    config = PlatformConfig()
+    config.technical.disabled_indicators = [
+        "ema_trend", "rsi", "bollinger", "macd", "adx",
+        "keltner", "roc", "candlestick", "divergence",
+    ]
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=42)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    assert report is not None
+    assert report.signal_quality in ("A+", "A", "B", "C", "D", "F", "")
+
+
+@test("Integration: data with 50% NaN still produces valid report")
+def _():
+    df = make_ohlcv(n=300, seed=42)
+    mask = np.random.RandomState(99).choice([True, False], size=len(df), p=[0.5, 0.5])
+    df.loc[mask, "Close"] = np.nan
+    df.loc[mask, "High"] = np.nan
+    df.loc[mask, "Low"] = np.nan
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    dfs = {Timeframe.H1: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    assert report is not None
+    assert report.signal_quality in ("A+", "A", "B", "C", "D", "F", "")
+
+
+@test("Integration: XAU and EUR produce independent results")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    df = make_ohlcv(n=300, seed=42, trend=0.001)
+
+    xau_config = PlatformConfig()
+    xau_agg = SignalAggregator(xau_config)
+    xau_dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    xau_report = xau_agg.analyze_from_dataframes(symbol=XAUUSD, dfs=xau_dfs, use_news=False)
+
+    eur_config = eur_make_config(Timeframe.H1)
+    eur_agg = SignalAggregator(eur_config)
+    eur_dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    eur_report = eur_agg.analyze_from_dataframes(symbol=EURUSD, dfs=eur_dfs, use_news=False)
+
+    assert xau_report.display_name == "XAU/USD"
+    assert eur_report.display_name == "EUR/USD"
+
+
+@test("Integration: missing confirmation TFs still produces valid report")
+def _():
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    df = make_ohlcv(n=300, seed=42)
+    dfs = {Timeframe.H1: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    assert report is not None
+    assert report.mtf_agreement is False
+
+
+@test("Integration: extreme volatility data does not crash")
+def _():
+    n = 300
+    dates = pd.date_range(end=pd.Timestamp.now(tz="UTC"), periods=n, freq="h")
+    np.random.seed(42)
+    close = 100.0 + np.cumsum(np.random.randn(n) * 10.0)
+    close = np.maximum(close, 1.0)
+    df = pd.DataFrame({
+        "Open": close * 0.99, "High": close * 1.05,
+        "Low": close * 0.95, "Close": close,
+        "Volume": [5000.0]*n,
+    }, index=dates)
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    assert report is not None
+    assert -100 <= report.final_score <= 100
+
+
+@test("Integration: very short data (50 bars) produces valid report")
+def _():
+    df = make_ohlcv(n=50, seed=42)
+    config = PlatformConfig()
+    agg = SignalAggregator(config)
+    dfs = {Timeframe.H1: df}
+    report = agg.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    assert report is not None
+
+
+@test("Integration: multiple seeds produce deterministic results")
+def _():
+    df = make_ohlcv(n=300, seed=42)
+    config = PlatformConfig()
+    agg1 = SignalAggregator(config)
+    agg2 = SignalAggregator(config)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    r1 = agg1.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    r2 = agg2.analyze_from_dataframes(symbol=XAUUSD, dfs=dfs, use_news=False)
+    assert r1.final_score == r2.final_score
+    assert r1.final_signal == r2.final_signal
+    assert r1.signal_quality == r2.signal_quality
+
+
+@test("Integration: session filter blocks outside hours correctly")
+def _():
+    from pair_signals.eurusd import make_config as eur_make_config
+    config = eur_make_config(Timeframe.H1)
+    config.eur_strategy.session_hours = [8, 9, 10, 11, 12, 13, 14, 15, 16]
+    agg = SignalAggregator(config)
+    n = 300
+    end_time = pd.Timestamp("2026-05-18 03:00:00", tz="UTC")
+    dates = pd.date_range(end=end_time, periods=n, freq="h")
+    np.random.seed(42)
+    close = 1.08 + np.cumsum(np.random.randn(n) * 0.001 + 0.0002)
+    df = pd.DataFrame({
+        "Open": close - 0.0001, "High": close + 0.0005,
+        "Low": close - 0.0005, "Close": close, "Volume": [1000.0]*n,
+    }, index=dates)
+    dfs = {Timeframe.H1: df, Timeframe.M15: df, Timeframe.H4: df}
+    report = agg.analyze_from_dataframes(symbol=EURUSD, dfs=dfs, use_news=False)
+    bar_hour = dates[-1].hour
+    if bar_hour not in config.eur_strategy.session_hours and report.signal_quality != "":
+        assert report.signal_quality == "F", \
+            f"Hour {bar_hour} should be blocked, got {report.signal_quality}"
+
+
 # ══════════════════════════════════════════════════════════════════════
 #  SUMMARY
 # ══════════════════════════════════════════════════════════════════════