RK-A/task-69de7223f309a98be0007e09

T

RK-A 59f9fabfa0 Update main.py

2026-06-04 19:34:46 +00:00

__init__.py

Add __init__.py

2026-06-04 19:19:46 +00:00

agent.py

Update agent.py

2026-06-04 19:18:51 +00:00

deep_agent.py

Update deep_agent.py

2026-06-04 19:18:57 +00:00

LICENSE

Initial commit

2026-06-02 07:03:41 +00:00

main.py

Add main.py

2026-06-04 19:19:09 +00:00

README.md

Update README.md

2026-06-04 19:19:15 +00:00

requirements.txt

Update requirements.txt

2026-06-04 19:19:21 +00:00

tools.py

Update tools.py

2026-06-04 19:19:28 +00:00

vfs.py

Add vfs.py

2026-06-04 19:19:34 +00:00

virtual_fs.py

Update virtual_fs.py

2026-06-04 19:19:40 +00:00

README.md

Deep Agent from Scratch

This repository contains a minimal implementation of a deep agent that can:

Search the web using DuckDuckGo.
Create virtual files in memory.
Export the virtual files to the real file system.

The agent is built using the new LangChain 1.x and LangGraph 1.x APIs.

Prerequisites

Python 3.11 or newer
Ollama running locally with a model such as llama3.1
pip install -r requirements.txt

Running the Agent

python agent.py

The script will run the agent with a sample prompt, create a virtual file report.txt, and export it to the exported_files directory.

File Structure

agent.py – Main implementation.
requirements.txt – Python dependencies.
README.md – This file.