Update agent.py

agent.py

@@ -3,15 +3,26 @@ from langchain.agents import create_agent
 from langchain.tools import tool
 from pydantic import SecretStr
 import re
-import json
+import sys
 
-# Initialize main LLM
+# Utility to initialize LLM with error handling
-llm = ChatOpenAI(
+
-    base_url="http://localhost:11434/v1",
+def init_llm(base_url: str, model: str, api_key: str = "ollama") -> ChatOpenAI:
-    api_key=SecretStr("ollama"),
+    try:
-    model="<название модели в LM Studio>",
+        return ChatOpenAI(
-    temperature=0.7,
+            base_url=base_url,
-)
+            api_key=SecretStr(api_key),
+            model=model,
+            temperature=0.7,
+        )
+    except Exception as e:
+        print(f"Ошибка при подключении к LLM: {e}")
+        sys.exit(1)
+
+# Main LLM
+LLM_URL = "http://localhost:1234/v1"
+LLM_MODEL = "llama3"
+llm = init_llm(LLM_URL, LLM_MODEL)
 
 # Define get_price tool with subagent
 @tool(name="get_price", description="Get price for a product in a city.")
@@ -20,12 +31,7 @@ def get_price(product: str, city: str) -> str:
     For demonstration it returns a static price string, but the structure
     mirrors a real subagent that could call another LLM.
     """
-    sub_llm = ChatOpenAI(
+    sub_llm = init_llm(LLM_URL, LLM_MODEL, api_key="ollama")
-        base_url="http://localhost:11434/v1",
-        api_key=SecretStr("ollama"),
-        model="<название модели в LM Studio>",
-        temperature=0.2,
-    )
     sub_agent = create_agent(
         llm=sub_llm,
         tools=[],
@@ -37,8 +43,8 @@ def get_price(product: str, city: str) -> str:
             for msg in response["messages"]:
                 if msg.get("role") == "assistant" and msg.get("content"):
                     return msg["content"].strip()
-    except Exception:
+    except Exception as e:
-        pass
+        print(f"Ошибка при вызове subagent: {e}")
     return "price: 100, store: SuperMarket"