2. Add tools, agent with tools created and functionnal

+ Small clean up on main.py

.env.template

@@ -1 +1,6 @@
+# LLM Parameters
 GOOGLE_API_KEY=your_google_api_key_here
+GRPC_VERBOSITY=NONE # Options: DEBUG, INFO, ERROR, NONE - to remove errors
+
+# Tools parameters
+TAVILY_API_KEY=     # Search engine 

README.md

@@ -6,6 +6,6 @@ Step 1 : follow the LangGraph tutorial "Build a custom workflow"
 Step 2 : use the base to create a different custom workflow
 
 To use, install all dependancies : requirements.txt
-Add your Google Gemini API key in .env
+Fullfil the .env (all parameters are mandatory)
 Run : python -m main
 Exit with "quit", "exit", "q"

main.py

@@ -1,19 +1,30 @@
-# from src.graph_builder import build_graph
-from src.graph_builder import graph_builder
-from src.nodes import chatbot
+# imports from langgraph
 from langgraph.graph import START, END
 
+# imports from local files
+from src.graph_builder import graph_builder
+from src.nodes import chatbot, BasicToolNode
+from src.utils import route_tools
+from src.tools import tool
 
-# The first argument is the unique node name
-# The second argument is the function or object that will be called whenever
-# the node is used.
 
-if __name__ == "__main__":
-    graph_builder.add_node("chatbot", chatbot)
-    graph_builder.add_edge(START, "chatbot")
-    graph_builder.add_edge("chatbot", END)
+def main():
+    # The first argument is the unique node name
+    # The second argument is the function or object that will be called whenever
+    # the node is used.
+    graph_builder.add_node("chatbot", chatbot)      # Create the chatbot node
+    graph_builder.add_edge("chatbot", END)          # Add the edge from chatbot node to END
+    tool_node = BasicToolNode(tools=[tool])         # Intermediate variable creation to create a tool node with the tools
+    graph_builder.add_node("tools", tool_node)      # Create a tool node with the tools
+    graph_builder.add_conditional_edges(            # Create the conditionnal edges from chatbot to tools
+        "chatbot",
+        route_tools,
+        {"tools": "tools", END: END},
+        )
+    graph_builder.add_edge("tools", "chatbot")      # Add the edge from tools node back to chatbot
+    graph_builder.add_edge(START, "chatbot")        # Add the edge from START to chatbot node
     graph = graph_builder.compile()
-
+    
     def stream_graph_updates(user_input: str):
         for event in graph.stream({"messages": [{"role": "user", "content": user_input}]}):
             for value in event.values():
@@ -32,4 +43,7 @@ if __name__ == "__main__":
             user_input = "What do you know about LangGraph?"
             print("User: " + user_input)
             stream_graph_updates(user_input)
-            break
+            break
+
+if __name__ == "__main__":
+    main()

src/graph_builder.py

@@ -17,16 +17,3 @@ class State(TypedDict):
 
 graph_builder = StateGraph(State)
 
-
-
-
-
-# from langgraph import Graph
-# from src.nodes import CustomNode
-
-# def build_graph():
-#     graph = Graph()
-#     # Add nodes and edges here
-#     node = CustomNode()
-#     graph.add_node(node)
-#     return graph

src/nodes.py

@@ -1,21 +1,47 @@
-import os
+# General imports
+import os, json
 from dotenv import load_dotenv
+
+# LangGraph imports
 from langchain.chat_models import init_chat_model
 from langgraph.graph import StateGraph
+from langchain_core.messages import ToolMessage
+
+# Local imports
 from src.graph_builder import State
+from src.tools import tools
 
 load_dotenv()
 
-#os.environ["GOOGLE_API_KEY"] = "..."
-
 llm = init_chat_model("google_genai:gemini-2.0-flash")
 
+# Modification: tell the LLM which tools it can call
+llm_with_tools = llm.bind_tools(tools)
+
 def chatbot(state: State):
-    return {"messages": [llm.invoke(state["messages"])]}
+    return {"messages": [llm_with_tools.invoke(state["messages"])]}
 
+class BasicToolNode:
+    """A node that runs the tools requested in the last AIMessage."""
 
-# class CustomNode:
-#     def __init__(self):
-#         pass
-#     def run(self):
-#         print("Custom node running")
+    def __init__(self, tools: list) -> None:
+        self.tools_by_name = {tool.name: tool for tool in tools}
+
+    def __call__(self, inputs: dict):
+        if messages := inputs.get("messages", []):
+            message = messages[-1]
+        else:
+            raise ValueError("No message found in input")
+        outputs = []
+        for tool_call in message.tool_calls:
+            tool_result = self.tools_by_name[tool_call["name"]].invoke(
+                tool_call["args"]
+            )
+            outputs.append(
+                ToolMessage(
+                    content=json.dumps(tool_result),
+                    name=tool_call["name"],
+                    tool_call_id=tool_call["id"],
+                )
+            )
+        return {"messages": outputs}

src/tools.py

@@ -0,0 +1,12 @@
+# General imports
+import os, json
+from dotenv import load_dotenv
+
+# LangGraph imports
+from langchain_tavily import TavilySearch
+
+load_dotenv()
+
+tool = TavilySearch(max_results=10)
+tools = [tool]
+

src/utils.py

@@ -1 +1,20 @@
+from src.graph_builder import State, START, END
+
 # Utility functions for LangGraph project
+
+def route_tools(                    # function that checks for tool_calls in the chatbot's output
+    state: State,
+):
+    """
+    Use in the conditional_edge to route to the ToolNode if the last message
+    has tool calls. Otherwise, route to the end.
+    """
+    if isinstance(state, list):
+        ai_message = state[-1]
+    elif messages := state.get("messages", []):
+        ai_message = messages[-1]
+    else:
+        raise ValueError(f"No messages found in input state to tool_edge: {state}")
+    if hasattr(ai_message, "tool_calls") and len(ai_message.tool_calls) > 0:
+        return "tools"
+    return END