Skip to main content

ChatGPT plugin

OpenAI plugins connect ChatGPT to third-party applications. These plugins enable ChatGPT to interact with APIs defined by developers, enhancing ChatGPT's capabilities and allowing it to perform a wide range of actions.

Plugins allow ChatGPT to do things like:

  • Retrieve real-time information; e.g., sports scores, stock prices, the latest news, etc.
  • Retrieve knowledge-base information; e.g., company docs, personal notes, etc.
  • Perform actions on behalf of the user; e.g., booking a flight, ordering food, etc.

This notebook shows how to use the ChatGPT Retriever Plugin within LangChain.

# STEP 1: Load

# Load documents using LangChain's DocumentLoaders
# This is from

from langchain_community.document_loaders import CSVLoader
from langchain_core.documents import Document

loader = CSVLoader(
data = loader.load()

# STEP 2: Convert

# Convert Document to format expected by
import json
from typing import List

def write_json(path: str, documents: List[Document]) -> None:
results = [{"text": doc.page_content} for doc in documents]
with open(path, "w") as f:
json.dump(results, f, indent=2)

write_json("foo.json", data)

# STEP 3: Use

# Ingest this as you would any other json file in
API Reference:CSVLoader | Document

Using the ChatGPT Retriever Pluginโ€‹

Okay, so we've created the ChatGPT Retriever Plugin, but how do we actually use it?

The below code walks through how to do that.

We want to use ChatGPTPluginRetriever so we have to get the OpenAI API Key.

import getpass
import os

os.environ["OPENAI_API_KEY"] = getpass.getpass("OpenAI API Key:")
OpenAI API Key: ยทยทยทยทยทยทยทยท
from langchain_community.retrievers import (
retriever = ChatGPTPluginRetriever(url="", bearer_token="foo")
retriever.invoke("alice's phone number")
[Document(page_content="This is Alice's phone number: 123-456-7890", lookup_str='', metadata={'id': '456_0', 'metadata': {'source': 'email', 'source_id': '567', 'url': None, 'created_at': '1609592400.0', 'author': 'Alice', 'document_id': '456'}, 'embedding': None, 'score': 0.925571561}, lookup_index=0),
Document(page_content='This is a document about something', lookup_str='', metadata={'id': '123_0', 'metadata': {'source': 'file', 'source_id': '', 'url': '', 'created_at': '1609502400.0', 'author': 'Alice', 'document_id': '123'}, 'embedding': None, 'score': 0.6987589}, lookup_index=0),
Document(page_content='Team: Angels "Payroll (millions)": 154.49 "Wins": 89', lookup_str='', metadata={'id': '59c2c0c1-ae3f-4272-a1da-f44a723ea631_0', 'metadata': {'source': None, 'source_id': None, 'url': None, 'created_at': None, 'author': None, 'document_id': '59c2c0c1-ae3f-4272-a1da-f44a723ea631'}, 'embedding': None, 'score': 0.697888613}, lookup_index=0)]

Was this page helpful?

You can leave detailed feedback on GitHub.