This is an example of retrieving and executing prompts from LangChain Hub.
LangChain Hub is a repository that collects prompts frequently used across various projects. This enables developers to efficiently search for, retrieve, and execute these prompts whenever needed, thereby streamlining their workflow.
Prompt Search and Categorization: Developers can easily find the desired prompts using keyword-based search and categorization.
Reusability: Once created, a prompt can be reused across multiple projects, reducing development time.
Real-time Execution: Retrieved prompts can be executed immediately through LangChain to view the results in real time.
Extensibility and Customization: In addition to the default prompts provided, users have the flexibility to add and modify prompts according to their needs.
Table of Contents
References
Environment Setup
[Note]
You can check LangChain Hub prompts at the address below.
You can retrieve prompts by using the prompt repo ID, and you can also get prompts for specific versions by adding the commit ID.
langchain-opentutorial is a package that provides a set of easy-to-use environment setup, useful functions and utilities for tutorials.
You can check LangChain Hub prompts at the address below.
You can retrieve prompts using the prompt repo ID, and you can also get prompts for specific versions by adding the commit ID.
# Set environment variables
from langchain_opentutorial import set_env
set_env(
{
"OPENAI_API_KEY": "",
# Get an API key for your Personal organization if you have not yet. The hub will not work with your non-personal organization's api key!
# If you already have LANGCHAIN_API_KEY set to a personal organization’s api key from LangSmith, you can skip this.
"LANGCHAIN_API_KEY": "",
"LANGCHAIN_TRACING_V2": "true",
"LANGCHAIN_ENDPOINT": "https://api.smith.langchain.com",
"LANGCHAIN_PROJECT": "Personal Prompts for LangChain",
}
)
Environment variables have been set successfully.
Getting Prompts from Hub
Retrieve and execute prompts directly from LangChain Hub to accelerate your workflow.
How to seamlessly integrate available prompts into your projects.
from langchain import hub
# Get the latest version of the prompt
prompt = hub.pull("rlm/rag-prompt")
# Print the prompt content
print(prompt)
input_variables=['context', 'question'] input_types={} partial_variables={} metadata={'lc_hub_owner': 'rlm', 'lc_hub_repo': 'rag-prompt', 'lc_hub_commit_hash': '50442af133e61576e74536c6556cefe1fac147cad032f4377b60c436e6cdcb6e'} messages=[HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=['context', 'question'], input_types={}, partial_variables={}, template="You are an assistant for question-answering tasks. Use the following pieces of retrieved context to answer the question. If you don't know the answer, just say that you don't know. Use three sentences maximum and keep the answer concise.\nQuestion: {question} \nContext: {context} \nAnswer:"), additional_kwargs={})]
# To get a specific version of prompt, specify the version hash
prompt = hub.pull("rlm/rag-prompt:50442af1")
prompt
ChatPromptTemplate(input_variables=['context', 'question'], input_types={}, partial_variables={}, metadata={'lc_hub_owner': 'rlm', 'lc_hub_repo': 'rag-prompt', 'lc_hub_commit_hash': '50442af133e61576e74536c6556cefe1fac147cad032f4377b60c436e6cdcb6e'}, messages=[HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=['context', 'question'], input_types={}, partial_variables={}, template="You are an assistant for question-answering tasks. Use the following pieces of retrieved context to answer the question. If you don't know the answer, just say that you don't know. Use three sentences maximum and keep the answer concise.\nQuestion: {question} \nContext: {context} \nAnswer:"), additional_kwargs={})])
Register Your Own Prompt to Prompt Hub
Registering your own prompt to Prompt Hub allows developers to share custom prompts with the community, making them reusable across various projects.
This feature enhances prompt standardization and efficient management, streamlining development and fostering collaboration.
from langchain.prompts import ChatPromptTemplate
prompt = ChatPromptTemplate.from_template(
"Summarize the following text based on the given content. Please write the answer in Korean\n\nCONTEXT: {context}\n\nSUMMARY:"
)
prompt
ChatPromptTemplate(input_variables=['context'], input_types={}, partial_variables={}, messages=[HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=['context'], input_types={}, partial_variables={}, template='Summarize the following text based on the given content. Please write the answer in Korean\n\nCONTEXT: {context}\n\nSUMMARY:'), additional_kwargs={})])
from langchain import hub
# Upload the prompt to the hub
hub.push("cjlee/simple-summary-korean-1", prompt)
The following is the output after successfully uploading to Hub.
ID/PromptName/Hash
from langchain import hub
# Get the prompt from the hub
pulled_prompt = hub.pull("teddynote/simple-summary-korean")
# Print the prompt content
print(pulled_prompt)
input_variables=['context'] input_types={} partial_variables={} metadata={'lc_hub_owner': 'teddynote', 'lc_hub_repo': 'simple-summary-korean', 'lc_hub_commit_hash': 'b7e31df5666de7758d72fd038875973520d141548280185ee5b5ba846f015308'} messages=[HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=['context'], input_types={}, partial_variables={}, template='주어진 내용을 바탕으로 다음 문장을 요약하세요. 답변은 반드시 한글로 작성하세요\n\nCONTEXT: {context}\n\nSUMMARY:'), additional_kwargs={})]
Set up the environment. You may refer to for more details.