zhenghongpeng / amazon-bedrock-custom-langchain-agent

Learn to build custom prompts and tools for LangChain agents

Amazon Bedrock Custom LangChain Agent

Create a custom LangChain agent dubbed "Agent AWS" that queries the AWS Well-Architected Framework and deploys Lambda functions, all backed by Amazon Bedrock and housed in a Streamlit chatbot.

Prerequisites

Before you begin, make sure you have the following:

• AWS Account: You'll need an AWS account with access to Claude via Amazon Bedrock.
• Lambda Role: Create an IAM role with Lambda execution permissions.
• S3 Bucket: Create an S3 bucket to store assets like Python deployment zips for Lambda.
• Python 3.9 or higher

Export these in your terminal:

export LAMBDA_ROLE=arn:aws:iam::ACCOUNT_ID:role/YourLambdaExecutionRole  
export S3_BUCKET=your_s3_bucket_name_here

Installation

Clone this repository:

git clone https://github.com/build-on-aws/amazon-bedrock-custom-langchain-agent.git

Navigate into the project directory:

cd amazon-bedrock-custom-langchain-agent

Install the required Python packages:

pip install -r requirements.txt

Usage

Run the Streamlit app:

streamlit run agent_aws_st.py

You can also run the agent in the terminal:

python agent_aws.py

Security

See CONTRIBUTING for more information.

License

This library is licensed under the MIT-0 License. See the LICENSE file.