Remove LLM mock fallbacks and surface real errors

agent/llm_client.py

@@ -87,7 +87,7 @@ class LLMClient:
         Returns response text or yields chunks if streaming
         """
         if not self.provider:
-            return self._mock_response(messages)
+            raise RuntimeError("No LLM provider configured. Set ANTHROPIC_MCP_1ST_BDAY or OPENAI_MCP_1ST_BDAY.")
         
         if self.provider == "openrouter":
             return self._call_openrouter(messages, temperature, max_tokens, stream)
@@ -104,8 +104,7 @@ class LLMClient:
     ) -> Iterator[str]:
         """Stream chat completion response"""
         if not self.provider:
-            yield self._mock_response(messages)
-            return
+            raise RuntimeError("No LLM provider configured. Set ANTHROPIC_MCP_1ST_BDAY or OPENAI_MCP_1ST_BDAY.")
         
         if self.provider == "openrouter":
             yield from self._stream_openrouter(messages, temperature, max_tokens)
@@ -159,7 +158,7 @@ class LLMClient:
             logger.error(f"OpenRouter API error: {e}")
             if monitor:
                 monitor.complete_call(call_id, success=False, error_message=str(e))
-            return self._mock_response(messages)
+            raise
     
     def _stream_openrouter(self, messages, temperature, max_tokens):
         """Stream from OpenRouter"""
@@ -217,7 +216,7 @@ class LLMClient:
             logger.error(f"OpenRouter streaming error: {e}")
             if monitor:
                 monitor.complete_call(call_id, success=False, error_message=str(e))
-            yield self._mock_response(messages)
+            raise
     
     def _call_openai(self, messages, temperature, max_tokens, stream):
         """Call OpenAI API"""
@@ -254,7 +253,7 @@ class LLMClient:
             logger.error(f"OpenAI API error: {e}")
             if monitor:
                 monitor.complete_call(call_id, success=False, error_message=str(e))
-            return self._mock_response(messages)
+            raise
     
     def _stream_openai(self, messages, temperature, max_tokens):
         """Stream from OpenAI"""
@@ -294,7 +293,7 @@ class LLMClient:
             logger.error(f"OpenAI streaming error: {e}")
             if monitor:
                 monitor.complete_call(call_id, success=False, error_message=str(e))
-            yield self._mock_response(messages)
+            raise
     
     def _call_anthropic(self, messages, temperature, max_tokens, stream):
         """Call Anthropic API"""
@@ -346,7 +345,7 @@ class LLMClient:
             if self.openai_key:
                 logger.info("Falling back to OpenAI due to Anthropic error")
                 return self._call_openai(messages, temperature, max_tokens, stream=False)
-            return self._mock_response(messages)
+            raise RuntimeError(f"Anthropic call failed: {err_msg}")
     
     def _stream_anthropic(self, messages, temperature, max_tokens):
         """Stream from Anthropic"""
@@ -401,26 +400,8 @@ class LLMClient:
                 logger.info("Falling back to OpenAI streaming due to Anthropic error")
                 yield from self._stream_openai(messages, temperature, max_tokens)
             else:
-                yield self._mock_response(messages)
+                raise RuntimeError(f"Anthropic streaming failed: {err_msg}")
     
     def _mock_response(self, messages: List[LLMMessage]) -> str:
-        """Return mock response when no API key available"""
-        last_user_msg = next((m.content for m in reversed(messages) if m.role == "user"), "")
-        
-        if "consultation" in last_user_msg.lower():
-            return json.dumps({
-                "questions": [
-                    "What is your total budget for this network deployment?",
-                    "Do you have any vendor preferences? (Cisco, Juniper, Arista, Ubiquiti, MikroTik)",
-                    "When do you need this network operational?",
-                    "Do you have existing infrastructure to integrate with?",
-                    "What are your bandwidth requirements per location?"
-                ],
-                "clarifications": [
-                    "How many total concurrent devices across all 3 locations?",
-                    "Do you need site-to-site VPN between locations?",
-                    "PCI-DSS compliance required for payment processing?"
-                ]
-            })
-        
-        return "Mock LLM response - please configure API keys"
+        """Deprecated: mocks disabled to avoid hiding real failures."""
+        raise RuntimeError("LLM mock responses are disabled. Provide a valid API key.")