Deep Agent

Experimental agentic-AI workbench — a LangChain demo notebook that combines Tavily web search with an OpenAI LLM into a research agent.

Contents (DEEP/)

File	Description
demoagent.ipynb	Step-by-step agent demo — web-search tool definition (Tavily), LLM wiring (OpenAI via langchain-openai), and the tool-calling loop.
main.py	Package scaffold entrypoint.
pyproject.toml, uv.lock	uv-managed project metadata.
requirements.txt	Runtime dependencies (langchain, deepagents, langchain-openai, tavily-python, ipykernel).

Setup

pip install -r requirements.txt      # or: uv sync

Set OPENAI_API_KEY and TAVILY_API_KEY in a .env file, then open DEEP/demoagent.ipynb.

License

No license specified — for learning/reference use.