Multi-Agent System (MAS) with LangGraph

Readings

1. The initial prompt: The initial prompt
2. Documentation: "A Multi-Agent System for Interrogative Decomposition of Human Inquiries Using LLM-Backed Sparse Dynamic Graphs"
3. Literature review: Literature review and research on related work

Setup

To set up the development environment, run the following commands:

uv venv --python 3.12
source .venv/bin/activate
uv sync

Add API Keys

export GEMINI_API_KEY="your-gemini-api-key"

export LANGCHAIN_TRACING_V2=true
export LANGCHAIN_API_KEY="your-langsmith-api-key"
export LANGCHAIN_PROJECT="mas-inquiry"
export LANGCHAIN_ENDPOINT="https://api.smith.langchain.com"

or write it in an .env file.

Unit Tests

PYTHONPATH=. uv run pytest tests/

CLI

python -m src.cli --query "Correct push-ups?"
python -m src.cli --query "Healthier options than burger and fries"
python -m src.cli --query "Research the current state of AI agents"