add basic code

.env

@@ -0,0 +1,4 @@
+LLM_PROVIDER=groq
+SUPABASE_URL=https://YOUR.supabase.co
+SUPABASE_SERVICE_KEY=your-supabase-service-key
+SYSTEM_PROMPT_PATH=system_prompt.txt

agent.py

@@ -0,0 +1,44 @@
+from config import settings
+from llm_provider import get_llm
+from tools import ALL_TOOLS
+from retrievers import custom_retriever
+from langgraph.graph import START, StateGraph, MessagesState
+from langgraph.prebuilt import tools_condition, ToolNode
+from langchain_core.messages import SystemMessage, HumanMessage
+
+# Load system prompt
+with open(settings.system_prompt_path, "r", encoding="utf-8") as f:
+    system_prompt = f.read()
+
+sys_msg = SystemMessage(content=system_prompt)
+
+def build_graph():
+    llm = get_llm(settings.llm_provider)
+    llm_with_tools = llm.bind_tools(ALL_TOOLS)
+    
+    def assistant(state: MessagesState):
+        return {"messages": [llm_with_tools.invoke(state["messages"])]}
+
+    def retriever(state: MessagesState):
+        similar_q = custom_retriever.retrieve(state["messages"][0].content)
+        example_msg = HumanMessage(content=f"Similar Q&A:\n\n{similar_q}")
+        return {"messages": [sys_msg] + state["messages"] + [example_msg]}
+
+    builder = StateGraph(MessagesState)
+    builder.add_node("retriever", retriever)
+    builder.add_node("assistant", assistant)
+    builder.add_node("tools", ToolNode(ALL_TOOLS))
+    builder.add_edge(START, "retriever")
+    builder.add_edge("retriever", "assistant")
+    builder.add_conditional_edges("assistant", tools_condition)
+    builder.add_edge("tools", "assistant")
+
+    return builder.compile()
+
+if __name__ == "__main__":
+    graph = build_graph()
+    question = input("Ask your question: ")
+    messages = [HumanMessage(content=question)]
+    results = graph.invoke({"messages": messages})
+    for m in results["messages"]:
+        print(m.content)

config.py

@@ -0,0 +1,12 @@
+import os
+from dotenv import load_dotenv
+
+load_dotenv()
+
+class Settings:
+    provider = os.getenv("LLM_PROVIDER", "groq")
+    supabase_url = os.getenv("SUPABASE_URL")
+    supabase_key = os.getenv("SUPABASE_SERVICE_KEY")
+    # Add other settings
+
+settings = Settings()

llm_provider.py

@@ -0,0 +1,19 @@
+from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain_groq import ChatGroq
+from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
+from config import settings
+
+def get_llm(provider: str):
+    if provider == "google":
+        return ChatGoogleGenerativeAI(model="gemini-2.0-flash", temperature=0)
+    elif provider == "groq":
+        return ChatGroq(model="qwen-qwq-32b", temperature=0)
+    elif provider == "huggingface":
+        return ChatHuggingFace(
+            llm=HuggingFaceEndpoint(
+                url="https://api-inference.huggingface.co/models/Meta-DeepLearning/llama-2-7b-chat-hf",
+                temperature=0,
+            ),
+        )
+    else:
+        raise ValueError(f"Unknown provider: {provider}")

requirements.txt

@@ -1,2 +1,18 @@
 gradio
-requests
+requests
+langchain
+langchain-community
+langchain-core
+langchain-google-genai
+langchain-huggingface
+langchain-groq
+langchain-tavily
+langchain-chroma
+langgraph
+huggingface_hub
+supabase
+arxiv
+pymupdf
+wikipedia
+pgvector
+python-dotenv

retrievers/__init__.py

@@ -0,0 +1,8 @@
+from .custom_retriever import retrieve
+
+class CustomRetriever:
+    @staticmethod
+    def retrieve(query):
+        return retrieve(query)
+
+custom_retriever = CustomRetriever()

retrievers/custom_retriever.py

@@ -0,0 +1,21 @@
+import os
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_community.vectorstores import SupabaseVectorStore
+from supabase.client import create_client
+from config import settings
+
+embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2")
+supabase = create_client(settings.supabase_url, settings.supabase_key)
+vector_store = SupabaseVectorStore(
+    client=supabase,
+    embedding=embeddings,
+    table_name="documents",
+    query_name="match_documents_langchain",
+)
+
+def retrieve(query: str) -> str:
+    results = vector_store.similarity_search(query)
+    if results:
+        return results[0].page_content
+    else:
+        return "No similar questions found."

tools/__init__.py

@@ -0,0 +1,7 @@
+# tools/__init__.py
+# Automatically gather all @tool-decorated tools for registration
+
+from .math_tools import add, multiply
+from .search_tools import wiki_search
+
+ALL_TOOLS = [add, multiply, wiki_search]

tools/math_tools.py

@@ -0,0 +1,45 @@
+from langchain_core.tools import tool
+
+@tool
+def add(a: float, b: float) -> float:
+    """Add two numbers."""
+    return a + b
+
+@tool
+def subtract(a: float, b: float) -> float:
+    """Subtract b from a."""
+    return a - b
+
+@tool
+def multiply(a: float, b: float) -> float:
+    """Multiply two numbers."""
+    return a * b
+
+@tool
+def divide(a: float, b: float) -> float:
+    """Divide a by b. Raises an error if b is zero."""
+    if b == 0:
+        raise ValueError("Cannot divide by zero.")
+    return a / b
+
+@tool
+def modulus(a: float, b: float) -> float:
+    """Return the modulus (remainder) of a divided by b."""
+    return a % b
+
+@tool
+def power(a: float, b: float) -> float:
+    """Return a raised to the power of b."""
+    return a ** b
+
+@tool
+def sqrt(x: float) -> float:
+    """Return the square root of x. Raises error if x is negative."""
+    if x < 0:
+        raise ValueError("Cannot compute the square root of a negative number.")
+    return x ** 0.5
+
+@tool
+def abs_val(x: float) -> float:
+    """Return the absolute value of x."""
+    return abs(x)

tools/search_tools.py

@@ -0,0 +1,33 @@
+from langchain_core.tools import tool
+from langchain_community.document_loaders import WikipediaLoader, ArxivLoader
+from langchain_community.tools.tavily_search import TavilySearchResults
+
+@tool
+def wiki_search(query: str) -> dict:
+    """Search Wikipedia for a query and return maximum 2 results."""
+    search_docs = WikipediaLoader(query=query, load_max_docs=2).load()
+    formatted = "\n\n---\n\n".join(
+        f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
+        for doc in search_docs
+    )
+    return {"wiki_results": formatted}
+
+@tool
+def web_search(query: str) -> dict:
+    """Search Tavily for a query and return maximum 3 results."""
+    search_docs = TavilySearchResults(max_results=3).invoke(query=query)
+    formatted = "\n\n---\n\n".join(
+        f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
+        for doc in search_docs
+    )
+    return {"web_results": formatted}
+
+@tool
+def arxiv_search(query: str) -> dict:
+    """Search Arxiv for a query and return maximum 3 results."""
+    search_docs = ArxivLoader(query=query, load_max_docs=3).load()
+    formatted = "\n\n---\n\n".join(
+        f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content[:1000]}\n</Document>'
+        for doc in search_docs
+    )
+    return {"arxiv_results": formatted}

tools/vector_tools.py

@@ -0,0 +1,8 @@
+from langchain_core.tools import tool
+# Example vector tool using retriever
+from retrievers import custom_retriever
+
+@tool
+def similar_question(query: str) -> str:
+    """Retrieve a similar question from the vector store."""
+    return custom_retriever.retrieve(query)