HomeLangChainUsing Arcade tools

Use LangGraph with Arcade

In this guide, let’s explore how to integrate Arcade tools into your LangGraph application. Follow the step-by-step instructions below.

Set up your environment

Install the required packages, and ensure your environment variables are set with your Arcade and OpenAI API keys:

pip install langchain-arcade langchain-openai langgraph

Configure API keys

Provide your Arcade and OpenAI API keys. You can store them in environment variables or directly in your code:

import os
 
arcade_api_key = os.environ.get("ARCADE_API_KEY", "YOUR_ARCADE_API_KEY")
openai_api_key = os.environ.get("OPENAI_API_KEY", "YOUR_OPENAI_API_KEY")

Create and manage Arcade tools

Use the ArcadeToolManager to retrieve specific tools or entire toolkits:

from langchain_arcade import ArcadeToolManager
 
manager = ArcadeToolManager(api_key=arcade_api_key)
 
# Fetch the "ScrapeUrl" tool from the "Web" toolkit
tools = manager.get_tools(tools=["Web.ScrapeUrl"])
print(manager.tools)
 
# Get all tools from the "Google" toolkit
tools = manager.get_tools(toolkits=["Google"])
print(manager.tools)

Set up the language model and memory

Create an AI model and bind your tools. Use MemorySaver for checkpointing:

from langchain_openai import ChatOpenAI
from langgraph.checkpoint.memory import MemorySaver
 
model = ChatOpenAI(model="gpt-4o", api_key=openai_api_key)
bound_model = model.bind_tools(tools)
 
memory = MemorySaver()

Create a ReAct-style agent

Use the prebuilt ReAct agent from LangGraph to handle your Arcade tools:

from langgraph.prebuilt import create_react_agent
 
graph = create_react_agent(model=bound_model, tools=tools, checkpointer=memory)

Provide configuration and user query

Supply a basic config dictionary and a user query. Notice that user_id is required for tool authorization:

config = {
    "configurable": {
        "thread_id": "1",
        "user_id": "[email protected]"
    }
}
user_input = {
    "messages": [
        ("user", "List any new and important emails in my inbox.")
    ]
}

Stream the response

Stream the assistant’s output. If the tool requires authorization, the agent may raise a NodeInterrupt:

from langgraph.errors import NodeInterrupt
 
try:
    for chunk in graph.stream(user_input, config, stream_mode="values"):
        chunk["messages"][-1].pretty_print()
except NodeInterrupt as exc:
    print(f"\nNodeInterrupt occurred: {exc}")
    print("Please authorize the tool or update the request, then re-run.")

Tips for selecting tools

  • Relevance: Pick only the tools you need. Avoid using all tools at once.
  • Avoid conflicts: Be mindful of duplicate or overlapping functionality.

Next steps

Now that you have integrated Arcade tools into your LangGraph agent, you can:

  • Experiment with different toolkits, such as “Math” or “Search.”
  • Customize the agent’s prompts for specific tasks.
  • Try out other language models and compare their performance.

Enjoy exploring Arcade and building powerful AI-enabled Python applications!

Generic authorizationUser authorization