🦜🕸️ Using Composio With LangGraph

Integrate Composio with LangGraph agents to let them seamlessly interact with external apps

Star A Repository on GitHub

In this example, we will use LangGraph Agent to star a repository on GitHub using Composio Tools

1. Install Packages

   Python

   pip install composio-langgraph
   
2. Import Libraries & Initialize ComposioToolSet

   Python

   from typing import Literal
   from langchain_openai import ChatOpenAI
   from langgraph.graph import MessagesState, StateGraph
   from langgraph.prebuilt import ToolNode
   from composio_langgraph import Action, ComposioToolSet, App
   
   composio_toolset = ComposioToolSet()
   
3. Connect Your GitHub Account

   You need to have an active GitHub Integration. Learn how to do this here

   CLI

   composio login
   composio add github
   

   Don’t forget to set your COMPOSIO_API_KEY and OPENAI_API_KEY in your environment variables.

4. Get And Bind Tools

   You can get all the tools for a given app as shown below, but you can get specific actions and filter actions using usecase & tags. Learn more here

   Python

   tools = composio_toolset.get_tools(
       apps=[App.GITHUB]
   )
   tool_node = ToolNode(tools)
   model = ChatOpenAI(temperature=0, streaming=True)
   model_with_tools = model.bind_tools(tools)
   
5. Define the model calling function

   Python

   def call_model(state: MessagesState):
       """
       Process messages through the LLM and return the response
       """
       messages = state["messages"]
       response = model_with_tools.invoke(messages)
       return {"messages": [response]}
   
6. Define the decision function for workflow routing

   Python

   def should_continue(state: MessagesState) -> Literal["tools", "__end__"]:
       """
       Determine if the conversation should continue to tools or end
       Returns:
           - "tools" if the last message contains tool calls
           - "__end__" otherwise
       """
       messages = state["messages"]
       last_message = messages[-1]
       if last_message.tool_calls:
           return "tools"
       return "__end__"
   
7. Define the workflow graph

   Python

   workflow = StateGraph(MessagesState)
   
   workflow.add_node("agent", call_model)
   workflow.add_node("tools", tool_node)
   workflow.add_edge("__start__", "agent")
   workflow.add_conditional_edges(
       "agent",
       should_continue,
   )
   workflow.add_edge("tools", "agent")
   
   app = workflow.compile()
   
8. Execute the workflow

   Python

   for chunk in app.stream(
       {
           "messages": [
               (
                   "human",
                   "Star the GitHub Repository composiohq/composio",
               )
           ]
       },
       stream_mode="values",
   ):
       chunk["messages"][-1].pretty_print()