diff --git a/apps/api/src/services/ai_router.py b/apps/api/src/services/ai_router.py
index 2035006e..b112799f 100644
--- a/apps/api/src/services/ai_router.py
+++ b/apps/api/src/services/ai_router.py
@@ -71,8 +71,12 @@ class AIProviderEnum(Enum):
     OLLAMA = "ollama"
     GEMINI = "gemini"
     CLAUDE = "claude"
-    # 2026-03-29 ogt: ADR-036 Nemotron Tool Calling (83.3% 精準度)
-    NVIDIA = "nvidia"
+    # 2026-04-02 ogt: C1 修復 — 對齊 Registry 實際名稱
+    # OpenClawNemoProvider.name = "openclaw_nemo" (一般推理, via .188)
+    # NemotronProvider.name = "nemotron" (Tool Calling, direct NVIDIA NIM)
+    # 舊版 NVIDIA = "nvidia" 已移除: Registry 無此 Provider
+    OPENCLAW_NEMO = "openclaw_nemo"
+    NEMOTRON = "nemotron"
 
 
 # Provider 對應延遲預算 (ms)
@@ -80,8 +84,9 @@ PROVIDER_LATENCY_BUDGET: dict[AIProviderEnum, int] = {
     AIProviderEnum.OLLAMA: 60000,  # 本地，允許較長處理時間
     AIProviderEnum.GEMINI: 30000,  # 雲端，較低延遲
     AIProviderEnum.CLAUDE: 30000,  # 雲端，較低延遲
-    # 2026-03-29 ogt: ADR-036 Nemotron Tool Calling (延遲 11-45s)
-    AIProviderEnum.NVIDIA: 60000,  # Tool Calling 專用，允許較長時間
+    # 2026-04-02 ogt: C1 修復 — 對齊 Registry 名稱
+    AIProviderEnum.OPENCLAW_NEMO: 60000,  # via .188 → NVIDIA NIM，允許較長時間
+    AIProviderEnum.NEMOTRON: 60000,  # Tool Calling 專用，允許較長時間
 }
 
 
@@ -210,21 +215,25 @@ class AIRouter:
         self._ollama_summary = self._model_registry.get_model("ollama", "summary")
         self._gemini_default = self._model_registry.get_model("gemini", "default")
         self._claude_default = self._model_registry.get_model("claude", "default")
-        # 2026-03-29 ogt: ADR-036 Nemotron Tool Calling
-        self._nvidia_default = self._model_registry.get_model("nvidia", "default")
+        # 2026-04-02 ogt: C1 修復 — openclaw_nemo (一般推理) + nemotron (Tool Calling)
+        self._openclaw_nemo_default = self._model_registry.get_model("nvidia", "default")
+        self._nemotron_default = self._model_registry.get_model("nvidia", "default")
+        # 向後相容別名
+        self._nvidia_default = self._openclaw_nemo_default
 
         # Provider 對應模型映射
         self._provider_models: dict[AIProviderEnum, str] = {
             AIProviderEnum.OLLAMA: self._ollama_default,
             AIProviderEnum.GEMINI: self._gemini_default,
             AIProviderEnum.CLAUDE: self._claude_default,
-            AIProviderEnum.NVIDIA: self._nvidia_default,  # ADR-036
+            AIProviderEnum.OPENCLAW_NEMO: self._openclaw_nemo_default,
+            AIProviderEnum.NEMOTRON: self._nemotron_default,
         }
 
         # 完整 Fallback 鏈 (Provider, Model)
-        # 2026-03-30 ogt: NVIDIA 成為首選仲裁，加入 Fallback 鏈首位
+        # 2026-04-02 ogt: C1 修復 — OPENCLAW_NEMO 首選仲裁
         self._full_fallback_chain: list[tuple[AIProviderEnum, str]] = [
-            (AIProviderEnum.NVIDIA, self._nvidia_default),
+            (AIProviderEnum.OPENCLAW_NEMO, self._openclaw_nemo_default),
             (AIProviderEnum.GEMINI, self._gemini_default),
             (AIProviderEnum.CLAUDE, self._claude_default),
             (AIProviderEnum.OLLAMA, self._ollama_default),
@@ -232,7 +241,7 @@ class AIRouter:
 
         # Tool Calling 專用 Fallback 鏈 (ADR-036)
         self._tool_calling_fallback_chain: list[tuple[AIProviderEnum, str]] = [
-            (AIProviderEnum.NVIDIA, self._nvidia_default),
+            (AIProviderEnum.NEMOTRON, self._nemotron_default),
             (AIProviderEnum.GEMINI, self._gemini_default),
             (AIProviderEnum.CLAUDE, self._claude_default),
         ]
@@ -394,12 +403,13 @@ class AIRouter:
             return provider, model, reason
 
         # =======================================================================
-        # 規則 4: 複雜度 4-5 或 HIGH 風險 → Nvidia Nemotron
+        # 規則 4: 複雜度 4-5 或 HIGH 風險 → OpenClaw Nemo (via .188 → NVIDIA NIM)
+        # 2026-04-02 ogt: C1 修復 — NVIDIA→OPENCLAW_NEMO 對齊 Registry 名稱
         # =======================================================================
         if score >= 4 or risk == RiskLevel.HIGH:
-            provider = AIProviderEnum.NVIDIA
-            model = self._nvidia_default
-            reason = f"複雜度={score}/5, 風險={risk.value} → Nvidia (fallback Gemini)"
+            provider = AIProviderEnum.OPENCLAW_NEMO
+            model = self._openclaw_nemo_default
+            reason = f"複雜度={score}/5, 風險={risk.value} → OpenClaw Nemo (fallback Gemini)"
             return provider, model, reason
 
         # =======================================================================
@@ -538,8 +548,9 @@ class AIRouter:
         Returns:
             (provider, model, fallback_chain)
         """
-        provider = AIProviderEnum.NVIDIA
-        model = self._nvidia_default
+        # 2026-04-02 ogt: C1 修復 — Tool Calling 使用 NEMOTRON (direct NIM)
+        provider = AIProviderEnum.NEMOTRON
+        model = self._nemotron_default
         fallback_chain = [
             (p, m) for p, m in self._tool_calling_fallback_chain if p != provider
         ]
@@ -599,8 +610,8 @@ class AIRouter:
             {
                 "rule": 4,
                 "condition": "complexity >= 4 OR HIGH risk",
-                "provider": "nvidia",
-                "reason": "高複雜度需要 Nvidia Nemotron 強大推理能力",
+                "provider": "openclaw_nemo",
+                "reason": "高複雜度需要 Nvidia Nemotron 強大推理能力 (via .188)",
             },
             {
                 "rule": 5,
@@ -852,8 +863,8 @@ class AIRouterExecutor:
                 continue
 
             # 閘門 2: Rate Limiter
-            # 2026-04-02 Claude Code: Phase 24 B3 — 加入 nemotron Rate Limiter
-            if provider_name in ("nvidia", "gemini", "claude", "nemotron"):
+            # 2026-04-02 Claude Code: Phase 24 B3 + C1 修復 — Rate Limiter (含 openclaw_nemo)
+            if provider_name in ("openclaw_nemo", "nemotron", "gemini", "claude"):
                 try:
                     from src.services.ai_rate_limiter import get_ai_rate_limiter
                     rate_limiter = get_ai_rate_limiter()