commit 3f3b6ec91d13f95782296b25dac17ae16c8502f6
Author: Глеб Никишин <glevelll@yandex.ru>
Date:   Thu May 28 17:37:08 2026 +0000

    add main.py

diff --git a/main.py b/main.py
new file mode 100644
index 0000000..89056a5
--- /dev/null
+++ b/main.py
@@ -0,0 +1,93 @@
+import os
+import sys
+from pathlib import Path
+from dotenv import load_dotenv
+
+from langchain_openai import ChatOpenAI
+from langchain_ollama import OllamaEmbeddings
+from langchain_chroma import Chroma
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+from langchain.tools import tool
+from langchain.agents import AgentExecutor, create_openai_tools_agent
+from langchain_community.tools.tavily import TavilySearchResults
+
+# Load env variables
+load_dotenv()
+
+# ---------- LLM and embeddings ----------
+llm = ChatOpenAI(
+    model="openai/gpt-oss-20b:free",
+    base_url="https://openrouter.ai/api/v1",
+    api_key=os.getenv("OPENAI_API_KEY"),
+    temperature=0.0,
+)
+embeddings = OllamaEmbeddings(model="nomic-embed-text")
+
+# ---------- Vectorstore ----------
+PERSIST_DIR = Path("./chroma_db")
+PERSIST_DIR.mkdir(parents=True, exist_ok=True)
+vectorstore = Chroma(persist_directory=str(PERSIST_DIR), embedding_function=embeddings)
+
+# Load documents if collection empty
+if not vectorstore.get_collection().count():
+    docs_dir = Path("./documents")
+    if docs_dir.exists():
+        splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+        for file in docs_dir.glob("**/*.*"):
+            if file.suffix.lower() in {".txt", ".md"}:
+                text = file.read_text(encoding="utf-8")
+                chunks = splitter.split_text(text)
+                vectorstore.add_texts(chunks, ids=[f"{file.name}_{i}" for i in range(len(chunks))])
+    vectorstore.persist()
+
+# ---------- Tools ----------
+@tool("search_local_kb")
+def search_local_kb(query: str, top_k: int = 3) -> str:
+    """Semantic search in local ChromaDB knowledge base."""
+    retriever = vectorstore.as_retriever(search_kwargs={"k": top_k})
+    docs = retriever.get_relevant_documents(query)
+    if not docs:
+        return "No relevant documents found in local KB."
+    return "\n---\n".join(doc.page_content for doc in docs)
+
+@tool("web_search")
+def web_search(query: str) -> str:
+    """Web search using Tavily."""
+    tavily = TavilySearchResults(max_results=3)
+    results = tavily.run(query)
+    if not results:
+        return "No web results found."
+    return "\n---\n".join(f"{r['title']}\n{r['url']}\n{r.get('content', '')}" for r in results)
+
+# ---------- Agent ----------
+SYSTEM_PROMPT = (
+    "You are an AI assistant. For questions about local documents, use the tool 'search_local_kb'. "
+    "For up-to-date information or news, use 'web_search'. Always indicate the source in your answer."
+)
+
+tools = [search_local_kb, web_search]
+agent = create_openai_tools_agent(llm, tools, system_message=SYSTEM_PROMPT)
+agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
+
+# ---------- CLI ----------
+
+def main():
+    print("RAG Agent with ChromaDB and Tavily. Type 'exit' to quit.")
+    while True:
+        try:
+            query = input("\nЗапрос: ")
+        except (EOFError, KeyboardInterrupt):
+            print("\nBye!")
+            break
+        if query.strip().lower() in {"exit", "quit"}:
+            print("Bye!")
+            break
+        result = agent_executor.invoke({"input": query})
+        # The tool name is stored in the tool_calls field of the result
+        tool_name = result.get("tool_calls", [{}])[0].get("name", "unknown")
+        source = "tavily" if tool_name == "web_search" else "chromadb"
+        print(f"\n[{'Web Search' if source=='tavily' else 'Local KB'}] {result.get('output', '')}\n")
+        print(f"Источник: {source}\n")
+
+if __name__ == "__main__":
+    main()