Skip to content

README.md

Latest commit

 

History

History
225 lines (162 loc) · 7.95 KB

README.md

File metadata and controls

225 lines (162 loc) · 7.95 KB
Raw

DevOps Agent 🤖

An intelligent AI-powered CLI assistant for DevOps engineers, providing expert guidance on Kubernetes, Terraform, and DevOps best practices. Built with multi-agent architecture and supporting multiple LLM providers.

✨ Features

  • Multi-Agent Architecture: Specialized agents for DevOps, Kubernetes, and Terraform domains
  • Multi-LLM Support: Choose between OpenAI, Anthropic (Claude), or Google (Gemini)
  • Interactive Mode: Conversational interface for continuous assistance
  • Log Analysis: Automated analysis of log files with insights and error detection
  • GitOps & Cloud-Native: Expert knowledge of modern DevOps practices
  • Memory System: Uses Qdrant vector database for contextual memory across sessions
  • POML Prompts: Structured prompts using POML (Prompt Markup Language) for consistent agent behavior

🚀 Quick Start

Installation

# Clone the repository
git clone https://github.com/yourusername/devops-agent.git
cd devops-agent

# Install dependencies
pip install -e .

Configuration

Create a .env file in the project root with your API keys:

# Choose one or more providers
ANTHROPIC_API_KEY=your_anthropic_key_here
OPENAI_API_KEY=your_openai_key_here
GEMINI_API_KEY=your_gemini_key_here

# Qdrant configuration (for memory)
QDRANT_URL=your_qdrant_url
QDRANT_API_KEY=your_qdrant_key

📖 Usage Examples

1. Interactive Mode with Anthropic (Recommended)

devops-agent run --provider anthropic --interactive

Start a conversation with the DevOps team. Ask multiple questions in sequence, and the agents will remember context across the session.

2. Single Query with OpenAI

devops-agent run --provider openai --query "How do I set up a multi-region Kubernetes cluster on AWS EKS with GitOps?"

Get immediate answers to specific DevOps questions without entering interactive mode.

3. Terraform Infrastructure Code Generation

devops-agent run --provider anthropic --query "Create a Terraform module for Azure AKS cluster with monitoring and auto-scaling" --output terraform-aks.tf --format text

Generate infrastructure code and save it directly to a file.

4. Kubernetes Troubleshooting

devops-agent run --provider google --query "My pods are in CrashLoopBackOff state. How do I debug this issue systematically?"

Get step-by-step troubleshooting guidance for Kubernetes issues.

5. Log File Analysis

devops-agent run --provider anthropic --log-file ./logs/application.log

Analyze log files for critical errors, patterns, anomalies, and significant findings with AI-powered insights.

6. Save Conversation to File

devops-agent run --provider openai --query "Best practices for implementing GitOps with ArgoCD" --output gitops-guide.md --format markdown

Export responses in text, JSON, or Markdown format for documentation.

🏗️ Project Structure

devops-agent/
├── devops_agent/
│   ├── cli.py                          # Main CLI interface with Click
│   ├── core/                           # Core agent implementations
│   │   ├── master_agent.py            # Orchestrator agent routing queries
│   │   ├── devops_agent.py            # DevOps troubleshooting specialist
│   │   ├── kubernetes_agent.py        # Kubernetes architecture expert
│   │   ├── terraform_agent.py         # Terraform/IaC specialist
│   │   └── log_analysis_agent.py      # Log file analysis agent
│   ├── prompts/                        # POML-based structured prompts
│   │   ├── devops.poml                # DevOps troubleshooting instructions
│   │   ├── kubernetes.poml            # Kubernetes architecture guidelines
│   │   └── terraform.poml             # Terraform best practices
│   └── utils/                          # Utility functions
│       └── prompt_generator_from_poml.py  # POML to prompt converter
├── setup.py                            # Package setup configuration
├── pyproject.toml                      # Modern Python project config
├── requirements.txt                    # Project dependencies
├── .env                                # Environment variables (create this)
└── README.md                           # This file

Key Components

  • cli.py: Command-line interface handling user input, interactive mode, and output formatting
  • master_agent.py: Routes queries to specialized agents based on domain (DevOps, K8s, Terraform)
  • devops_agent.py: Handles incident response, observability, and general DevOps troubleshooting
  • kubernetes_agent.py: Kubernetes architecture, GitOps workflows, and cloud-native practices
  • terraform_agent.py: Infrastructure as Code, state management, and multi-cloud deployments
  • log_analysis_agent.py: Analyzes log files for errors, patterns, and anomalies
  • .poml files: Structured prompt definitions using POML markup language

🛠️ CLI Commands

Run Command

devops-agent run [OPTIONS]

Options:
  --provider TEXT              LLM provider: openai, anthropic, or google
  --query TEXT                 Single query to process
  --log-file PATH              Path to log file for analysis
  --output PATH                Save response to file
  --format [text|json|markdown] Output format (default: text)
  -i, --interactive            Run in interactive conversation mode
  --help                       Show help message

Version

devops-agent --version

🧠 How It Works

  1. Query Routing: Master agent analyzes your question and routes it to the appropriate specialist agent
  2. Domain Expertise: Each agent uses specialized POML-defined knowledge for accurate responses
  3. Memory System: Qdrant vector database stores conversation history for contextual awareness
  4. Multi-LLM: Choose the best model for your needs - Claude Sonnet 4.5, GPT-5-mini, or Gemini 2.5
  5. Streaming Responses: See agent reasoning and responses in real-time

📋 Requirements

  • Python >= 3.8
  • API key for at least one LLM provider (OpenAI, Anthropic, or Google)
  • Optional: Qdrant instance for persistent memory (can use in-memory fallback)

🤝 Contributing

We welcome contributions! Here's how you can help:

  1. Fork the repository
  2. Create a feature branch: git checkout -b feature/amazing-feature
  3. Make your changes and commit: git commit -m 'Add amazing feature'
  4. Push to the branch: git push origin feature/amazing-feature
  5. Open a Pull Request

Development Setup

# Install with development dependencies
pip install -e ".[dev]"

# Format code
black devops_agent/
isort devops_agent/

# Lint
flake8 devops_agent/

Contribution Guidelines

  • Follow PEP 8 style guidelines
  • Add tests for new features
  • Update documentation as needed
  • Keep commits atomic and well-described
  • Ensure all tests pass before submitting PR

📝 License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

🙏 Acknowledgments

  • Built with Agno framework for multi-agent orchestration
  • Uses POML for structured prompt engineering
  • Powered by Claude (Anthropic), GPT (OpenAI), and Gemini (Google)

📬 Support

🗺️ Roadmap

  • Add CI/CD pipeline integration
  • Support for custom agent creation
  • Web UI interface
  • Docker and Helm chart templates generation
  • Integration with monitoring tools (Prometheus, Grafana)
  • Multi-language support for responses
  • Plugin system for extensibility

Made with ❤️ by M K Pavan Kumar