Agent AWS is an automated, AI-powered agent that uses HuggingFace Transformers paired with numerous different foundation models.
An “agent” here is a large language model (LLM) equipped with a prompt and access to a specific set of tools. Tools are self-contained functions that perform a singular task that the agent can use when given an instruction.
Agent AWS contains a set of tools that allows it to query AWS documentation, generating code, and creating architectural diagrams.
To learn more read the blog post here.
Here is how you can easily get started using the Agent
Checkout the code
git clone https://github.com/aws-banjo/building-aws-agent.git
cd building-aws-agent
Run the Agent
pip install -r requirements.txt
# Learn how to get your own free key here https://huggingface.co/docs/hub/security-tokens
export HUGGING_FACE_KEY=YOUR_KEY
streamlit run agent_aws_st.py
# View at localhost:8501
Here are some example outputs
