commit 11d3efd138c19c20a67b439f45a9edb1d6c9b2cf
Author: Danil Parunin 5f1b81b8-4f5d-11e8-9c2d-fa7ae01bbebc <dapa46@mail.ru>
Date:   Mon Jun 15 12:31:24 2026 +0000

    add: main.py

diff --git a/main.py b/main.py
new file mode 100644
index 0000000..a60cc37
--- /dev/null
+++ b/main.py
@@ -0,0 +1,75 @@
+import os
+import asyncio
+from langchain_openai import ChatOpenAI
+from langchain_core.messages import HumanMessage
+from langchain.tools import tool
+from deepagents import create_deep_agent
+from deepagents.backends import CompositeBackend, LocalShellBackend, FilesystemBackend
+
+# ---------- LLM ----------
+llm = ChatOpenAI(
+    model="openai/gpt-oss-20b:free",
+    base_url="https://openrouter.ai/api/v1",
+    api_key=os.getenv("OPENAI_API_KEY"),
+    temperature=0.7,
+)
+
+# ---------- Backend ----------
+backend = CompositeBackend([
+    LocalShellBackend(workspace_dir="./workspace"),
+    FilesystemBackend(),
+])
+
+# ---------- Sub‑agent for price generation ----------
+# The sub‑agent simply asks the LLM to produce a realistic price table.
+# It is wrapped in a tool so that the main agent can call it.
+
+@tool
+def get_price(product: str, city: str) -> str:
+    """Return a realistic price for a product in a given city.
+    The response must be a Markdown table with columns: Продукт, Цена (руб.), Магазин.
+    """
+    # Create a tiny agent that only generates the table.
+    from langchain.agents import create_agent
+    from langchain_core.messages import HumanMessage
+
+    system_prompt = (
+        "You are a market price generator. "
+        "Given a product and a city, produce a realistic price table in Markdown. "
+        "Use plausible Russian store names and prices."
+    )
+    sub_agent = create_agent(
+        model=llm,
+        tools=[],
+        system_prompt=system_prompt,
+    )
+    prompt = f"Product: {product}\nCity: {city}"
+    result = sub_agent.invoke({"messages": [HumanMessage(content=prompt)]})
+    # The sub‑agent returns a dict with 'messages'; take the last content.
+    return result["messages"][-1].content
+
+# ---------- Main agent ----------
+agent = create_deep_agent(
+    model=llm,
+    tools=[get_price],
+    backend=backend,
+    system_prompt="Ты помощник по планированию покупок.",
+)
+
+# ---------- Run ----------
+async def main():
+    user_query = "Помоги составить список покупок: молоко, хлеб, яблоки. Я нахожусь в Казани."
+    result = await agent.ainvoke(
+        {"messages": [HumanMessage(content=user_query)]},
+        {"configurable": {"thread_id": "session-1"}},
+    )
+    # Print all messages in order
+    for msg in result["messages"]:
+        if msg.content:
+            print(msg.content)
+        elif msg.tool_calls:
+            for call in msg.tool_calls:
+                print(f"{call['name']}({call['args']})")
+
+if __name__ == "__main__":
+    asyncio.run(main())