Automation Workflow

This repository implements an AI-driven automation workflow for software development, integrating Azure DevOps, Git, and Azure Portal. The workflow automates task retrieval, code generation, code review, and deployment.

Structure

• python-agent/: Python-based AI Agent for task processing, code generation, and Slack integration.
• mcp-servers/: Node.js-based MCP servers for Azure DevOps, Git, and Azure Portal.

Setup

1. Python Agent:

   • Navigate to python-agent/.
   • Install dependencies: pip install -r requirements.txt.
   • Configure environment variables: SLACK_TOKEN, AZURE_ORG, AZURE_PROJECT, AZURE_PAT.
   • Run: python src/main.py.

2. MCP Servers:

   • Navigate to each server folder (azure-devops-server, git-server, azure-portal-server).
   • Install dependencies: npm install.
   • Run each server: npm start.

Usage

1. The AI Agent fetches tasks from Azure DevOps via the MCP server.
2. Tasks are displayed in Slack.
3. The AI generates Express.js/React.js code based on selected tasks.
4. Code is committed to a Git branch and a PR is created.
5. Deployment is triggered to Azure App Service via the MCP server.

Requirements

• Python 3.8+
• Node.js 16+
• Slack account and bot token
• Azure DevOps PAT
• Azure CLI
• GitHub PAT