🦜️🔗 LangChain 🤝 Amazon Web Services (AWS)

This monorepo provides LangChain and LangGraph components for various AWS services. It aims to replace and expand upon the existing LangChain AWS components found in the langchain-community package in the LangChain repository.

The following packages are hosted in this repository:

• langchain-aws (PyPI)
• langgraph-checkpoint-aws (PyPI)
• langchain-agentcore-codeinterpreter (PyPI)

Features

LangChain

• LLMs: Includes LLM classes for AWS services like Bedrock and SageMaker Endpoints, allowing you to leverage their language models within LangChain.
• VectorStores: Supports vectorstores for services like Amazon MemoryDB, Amazon S3 Vectors, and AWS ElastiCache for Valkey, providing efficient and scalable vector database for your applications.
• Retrievers: Supports retrievers for services like Amazon Kendra and KnowledgeBases for Amazon Bedrock, enabling efficient retrieval of relevant information in your RAG applications.
• Graphs: Provides components for working with AWS Neptune graphs within LangChain.
• Agents: Includes Runnables to support Amazon Bedrock Agents, allowing you to leverage Bedrock Agents within LangChain and LangGraph.
• Tools: Includes tools and toolkits to enable use of Amazon Bedrock AgentCore's built-in tools with LangChain and LangGraph agents.

LangGraph

• Checkpointers: Provides custom checkpointing solutions for LangGraph agents using several AWS services, including Bedrock AgentCore Memory, Bedrock Session Management, DynamoDB, and ElastiCache Valkey.
• Memory Stores - Provides memory store solutions for saving, processing, and retrieving intelligent long term memories using services like Bedrock AgentCore Memory and ElastiCache Valkey.

Deep Agents

• Sandboxes: Provides an Amazon Bedrock AgentCore Code Interpreter sandbox backend for Deep Agents, enabling secure code execution in isolated MicroVM environments.

Note: This repository will replace all AWS integrations currently present in the langchain-community package. Users are encouraged to migrate to this repository as soon as possible.

Installation

You can install the langchain-aws package from PyPI.

pip install langchain-aws

The langgraph-checkpoint-aws package can also be installed from PyPI.

pip install langgraph-checkpoint-aws

The langchain-agentcore-codeinterpreter package can also be installed from PyPI.

pip install langchain-agentcore-codeinterpreter

Usage

langchain-aws

Here's a simple example of how to use the langchain-aws package.

from langchain_aws import ChatBedrockConverse
Initialize the Bedrock chat model
model = ChatBedrockConverse(
    model="us.anthropic.claude-sonnet-4-5-20250929-v1:0"
)
Invoke the model
response = model.invoke("Hello! How are you today?")
print(response)

AgentCore Tools

from langchain_aws.tools import create_browser_toolkit, create_code_interpreter_toolkit
Browser automation
browser_toolkit, browser_tools = create_browser_toolkit(region="us-west-2")
Code execution (async)
code_toolkit, code_tools = await create_code_interpreter_toolkit(region="us-west-2")
Use with LangGraph agent
agent = create_react_agent(model, tools=browser_tools + code_tools)
result = await agent.ainvoke(
    {"messages": [{"role": "user", "content": "Navigate to example.com"}]},
    config={"configurable": {"thread_id": "session-1"}}
)
Cleanup
await browser_toolkit.cleanup()
await code_toolkit.cleanup()

For more detailed usage examples and documentation, please refer to the LangChain docs.

[!TIP]

For developing, debugging, and deploying AI agents and LLM applications, see LangSmith.

langgraph-checkpoint-aws

You can find usage examples for langgraph-checkpoint-aws in the README.

langchain-agentcore-codeinterpreter

from bedrock_agentcore.tools.code_interpreter_client import CodeInterpreter
from langchain_agentcore_codeinterpreter import AgentCoreSandbox
interpreter = CodeInterpreter(region="us-west-2")
interpreter.start()
backend = AgentCoreSandbox(interpreter=interpreter)
result = backend.execute("echo hello")
print(result.output)  # hello
interpreter.stop()

Contributing

We welcome contributions to this repository! To get started, please follow the Contributing Guide.

This guide provides detailed instructions on how to set up each project for development and guidance on how to contribute effectively.

License

This project is licensed under the MIT License.