🚀 Production-ready AWS DevOps automation agent built with modern Python best practices, using Strands framework + Bedrock Agent Core deployment, integrating official AWS MCP servers for real-time optimization, compliance, and infrastructure management.
- ⚡ Quick Setup - Get the agent running in 3 minutes with step-by-step installation
- 📖 User Manual - Complete guide to using the AWS DevOps Agent with detailed examples and workflows
- 🚀 App Info - Comprehensive overview of features, architecture, and implementation status
This agent transforms AWS DevOps operations into intelligent, automated processes:
- Real-time cost optimization with AWS Pricing API integration
- Infrastructure as Code analysis for Terraform and CloudFormation
- Security and compliance validation (SOC2, HIPAA, PCI-DSS, ISO27001)
- Multi-account AWS operations across organizations
- GitHub MCP integration with repository management and PR automation
- Document generation with automatic report creation in organized folders
Strands Framework → Bedrock Agent Core
┌─────────────────────────┐ ┌─────────────────────────┐
│ AWS DevOps Agent │ → │ AWS Bedrock Runtime │
│ ├── Cost Optimization │ │ ├── Claude Sonnet 4 │
│ ├── IaC Analysis │ │ ├── HTTP API Service │
│ ├── Compliance Check │ │ └── /invocations │
│ ├── Multi-Account Mgmt │ └─────────────────────────┘
│ └── GitHub Automation │
└─────────────────────────┘
│
┌─────────▼──────────┐
│ AWS MCP Servers │
│ ├── Pricing │
│ ├── DynamoDB │
│ ├── Cost Explorer │
│ ├── Terraform │
│ ├── CloudWatch │
│ └── GitHub │
└────────────────────┘
strands-bedrock-mcp-devops-agent/
├── main.py # 🚀 Main entry point
├── pyproject.toml # 📦 Modern Python configuration
├── requirements.txt # 📋 Production dependencies
├── requirements_dev.txt # 🔧 Development dependencies
├── .env.example # 🔐 Environment template
│
├── src/aws_devops_agent/ # 📂 Main package (modern src/ layout)
│ ├── __init__.py
│ ├── main.py # Core agent implementation
│ ├── config/ # ⚙️ Configuration management
│ │ ├── __init__.py
│ │ └── app_config.py
│ ├── tools/ # 🛠️ Organized AWS DevOps tools
│ │ ├── __init__.py
│ │ ├── aws_cost/ # 💰 Cost optimization
│ │ │ ├── __init__.py
│ │ │ ├── pricing.py # AWS pricing analysis
│ │ │ ├── optimization.py # Cost Explorer integration
│ │ │ ├── resources.py # Live resource scanning
│ │ │ └── multi_account.py # Multi-account operations
│ │ ├── aws_iac/ # 🏗️ Infrastructure as Code
│ │ │ ├── __init__.py
│ │ │ └── terraform.py # Terraform/CloudFormation
│ │ ├── aws_compliance/ # 🔒 Security & compliance
│ │ │ ├── __init__.py
│ │ │ └── security.py # SOC2, HIPAA, PCI-DSS, ISO27001
│ │ └── github/ # 📱 GitHub integration
│ │ ├── __init__.py
│ │ └── integration.py # PR automation
│ └── mcp_clients/ # 🔌 MCP client integration
│ ├── __init__.py
│ ├── mcp_client.py # Unified MCP client
│ ├── aws_mcp_client.py # AWS-specific client
│ ├── strands_mcp_client.py # Strands-native client
│ └── github_mcp_client.py # GitHub MCP client
│
├── deployment/bedrock/ # 🚀 Production deployment
│ └── app.py # Bedrock Agent Core app
├── scripts/ # 🔧 Setup and utility scripts
│ └── setup.sh
├── tests/ # 🧪 Organized testing
│ ├── unit/ # Unit tests
│ ├── integration/ # Integration tests
│ └── fixtures/ # Test fixtures
└── docs/ # 📚 Documentation
├── APP_INFO.md
└── APP_USAGE.md # 📖 Comprehensive usage guide
📖 For detailed setup instructions, see Quick Setup Guide
- Python 3.10+
- AWS credentials configured
- Strands SDK access
- uv package manager
# Complete setup and run in one command
make setup
source .venv/bin/activate
make run# 1. Clone repository
git clone https://github.com/your-org/strands-bedrock-mcp-devops-agent
cd strands-bedrock-mcp-devops-agent
# 2. Create virtual environment
python3 -m venv .venv
source .venv/bin/activate
# 3. Install dependencies
pip install -r requirements.txt
# 4. Install MCP Servers
make mcp-install
# 5. Configure environment
cp .env.example .env
# Edit .env with your AWS credentials
# 6. Run the agent
PYTHONPATH=src python main.py --mode interactive# Main Commands
make run # Start interactive mode - MAIN COMMAND
make dev # Start demo mode
make query QUERY="your query" # Single query
# Setup & Installation
make setup # Complete setup: create venv, install deps, install MCP servers
make install # Alias for setup (same as make setup)
make clean # Clean up temporary files and virtual environment
# MCP Server Management
make mcp-check # Check if MCP servers are installed
make mcp-install # Install AWS MCP servers + GitHub MCP server
make mcp-test # Test MCP server connections
# Development & Testing
make test # Run core tests (unit + document generation)
make test-integration # Run integration tests (requires GitHub config)
make format # Format code
# Examples & Demos
make example TYPE=cost # Run cost analysis example
make example TYPE=iac # Run IaC analysis example👤 analyze
🤖 I'll analyze your AWS infrastructure costs using the Cost Explorer MCP Server...
📊 Cost Analysis Summary (Last 30 Days):
• Total Spend: $1,234.56
• Top Service: EC2 (37% - $456.78)
• Growth: +12.5% vs previous month
🎯 Top Recommendations:
1. Reserved Instances: Save up to $164/month on EC2
2. S3 Intelligent Tiering: Save 68% on infrequent access
3. Rightsize underutilized t3.large instances
💡 Quick Wins:
• Enable Cost Anomaly Detection
• Set up billing alerts at $1,500
• Review 8 idle EBS volumes ($23/month savings)
👤 compare t3.medium,m5.large for web application
🤖 I'll compare these instances for web application workloads...
💰 Cost Comparison (us-east-1, 24/7 usage):
t3.medium:
• On-Demand: $30.37/month
• Reserved (1yr): $19.54/month (36% savings)
• Best for: Variable web traffic, burst capability
m5.large:
• On-Demand: $70.08/month
• Reserved (1yr): $44.90/month (36% savings)
• Best for: Consistent high traffic, memory-intensive apps
🎯 Recommendation:
Start with t3.medium for cost efficiency. Upgrade to m5.large when:
• CPU utilization consistently >70%
• Memory usage >6GB consistently
• Response time SLA requirements increase
# Analyze costs across multiple AWS accounts
agent = DevOpsAgent()
await agent.analyze_infrastructure_costs(account_id="123456789012")# Compare pricing across regions
response = await agent.chat("""
Compare EC2 costs between us-east-1, eu-west-1, and ap-southeast-1
for a m5.large instance running 24/7. Include data transfer costs.
""")# Development mode
python main.py --mode interactive
# Run tests
pytest tests/
# Code formatting
black src/ tests/
isort src/ tests/
# Type checking
mypy src/# Install as package
pip install -e .
# Use CLI command
aws-devops-agent --mode interactive
# Bedrock Agent Core deployment
python deployment/bedrock/app.pyexport AWS_DEFAULT_REGION=us-east-1
export AWS_PROFILE=default
export BEDROCK_REGION=us-east-1{
"Version": "2012-10-17",
"Statement": [
{
"Effect": "Allow",
"Action": [
"ce:GetCostAndUsage",
"ce:GetForecast",
"cloudwatch:GetMetricData",
"cloudwatch:DescribeAlarms",
"logs:StartQuery",
"logs:GetQueryResults",
"pricing:GetProducts",
"pricing:DescribeServices"
],
"Resource": "*"
}
]
}- ✅ Strands SDK: Installed and functioning (v1.7.0)
- ✅ Bedrock Agent Core: Available (v0.1.2)
- ✅ MCP Protocol: Working (v1.13.1)
- ✅ AWS Configuration: Loaded correctly (us-east-1)
- ✅ All Tools: Cost, Compliance, Multi-Account, GitHub, IaC tools ready
# Run all tests
pytest tests/
# Run specific test categories
pytest tests/unit/
pytest tests/integration/
# Interactive testing
python tests/interactive_test.pyThis is a demonstration project for AWS re:Invent presentation. The architecture shows how to combine:
- Modern Python practices with industry standards
- Strands SDK for AI orchestration
- Official AWS MCP Servers for real-time data
- Bedrock Agent Core for sophisticated reasoning
- Clean architecture for maintainable DevOps operations
- Python 3.10+
- Strands SDK (requires access)
- AWS credentials configured
- uv package manager (installs automatically)
- 🎯 App Usage - Detailed usage examples, tool categories, and advanced workflows
- 🌍 Environment Variables - Complete configuration guide for all environment variables
- 🔒 Safety Guidelines - Critical security measures and safety protocols for production use
- 🚀 AgentCore Deployment Guide - Step-by-step Bedrock AgentCore deployment with safety measures
- ✅ Deployment Checklist - Production deployment verification checklist
- 📄 Document Generation - Automatic report generation and file organization
- 🛡️ AWS Security Analysis - Real-time security analysis using AWS Security Hub, Config, and Inspector
- 🏗️ Terraform Analysis - Comprehensive Terraform project analysis and optimization
- 📦 CDK Analysis - AWS Cloud Development Kit project analysis and cost optimization
- 🐙 GitHub MCP Integration - Complete GitHub integration for PR automation and repository management
- 🔧 GitHub MCP Quick Reference - Essential commands and quick setup for GitHub MCP server
- 📁 Demos - Interactive demo scripts and examples for testing agent capabilities
MIT License
Copyright (c) 2024 AWS DevOps Agent
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
