This Python project demonstrates an innovative approach to collaborative development using AI-driven agents. By leveraging the crewai
library alongside the langchain_community.llms
module, the project showcases how different roles within a software development team can be simulated and automated to perform tasks such as drafting user stories, coding, and code review.
- AI-Driven Agents: Utilizes
crewai
to create agents with specific roles and goals, including a Product Owner, Bash Scripting Expert, and Reviewer. - Language Model Integration: Leverages
Ollama
fromlangchain_community.llms
to loadMixtral-8x7B
local models, enabling sophisticated language understanding and generation capabilities. - Collaborative Task Management: Demonstrates how these agents can work together as an Agile team, following a sequential process to Define user stories, Build implementation and Review code.
- Ensure you have Python 3.11+ installed on your system:
conda create -n ai-agile-team python=3.11 -y
conda activate ai-agile-team
-
Install Ollama and Mixtral-8x7B model
- First install Ollama, see: https://ollama.com/download
- Then install Mixtral-8x7B model, run:
ollama run mixtral
-
Then, clone this repository and install the required dependencies:
git clone https://github.com/jeanjerome/ai-agile-team.git
cd ai-agile-team
pip install -r requirements.txt
To run the project and see the AI-driven development process in action:
conda activate ai-agile-team
python main.py
crewai
: A library for creating and managing AI agents in collaborative tasks.langchain_community.llms
: Provides access to language models likeOllama
for natural language understanding and generation.
We welcome contributions! If you have ideas for new features or improvements, please open an issue or submit a pull request.
This project is open source and available under the MIT License.