Polygon IO Toolkit

This notebook shows how to use agents to interact with the Polygon IO toolkit. The toolkit provides access to Polygon's Stock Market Data API.

Example Use

Setup

%pip install --upgrade --quiet langchain-community > /dev/null

Get your Polygon IO API key here, and then set it below. Note that the tool used in this example requires a "Stocks Advanced" subscription

import getpass
import os

os.environ["POLYGON_API_KEY"] = getpass.getpass()

········

It's also helpful (but not needed) to set up LangSmith for best-in-class observability

# os.environ["LANGCHAIN_TRACING_V2"] = "true"
# os.environ["LANGCHAIN_API_KEY"] = getpass.getpass()

Initializing the agent

from langchain import hub
from langchain.agents import AgentExecutor, create_openai_functions_agent
from langchain_community.agent_toolkits.polygon.toolkit import PolygonToolkit
from langchain_community.utilities.polygon import PolygonAPIWrapper
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(temperature=0)

instructions = """You are an assistant."""
base_prompt = hub.pull("langchain-ai/openai-functions-template")
prompt = base_prompt.partial(instructions=instructions)

API Reference:

• AgentExecutor
• create_openai_functions_agent
• PolygonToolkit
• PolygonAPIWrapper
• ChatOpenAI

polygon = PolygonAPIWrapper()
toolkit = PolygonToolkit.from_polygon_api_wrapper(polygon)
agent = create_openai_functions_agent(llm, toolkit.get_tools(), prompt)

agent_executor = AgentExecutor(
    agent=agent,
    tools=toolkit.get_tools(),
    verbose=True,
)

Get the last price quote for a stock

agent_executor.invoke({"input": "What is the latest stock price for AAPL?"})