PowerWorkflow is an open-source collection of agentic workflows for power system applications. These workflows enable intelligent automation and coordination of power system operations, facilitating efficient management and control in the energy domain.

Agentic workflows are step-by-step processes that process logic together. They provide a structured way to implement intelligent behavior without explicit agent programming. Key features include:

• Automated execution of complex power system operations
• Integration with various power system software and tools (PowerMCP)
• Reusable workflow components for different use cases
• Flexible orchestration of multiple tools and services

We're building an open-source community focused on advancing workflow automation in the power domain. Our goals are:

• Collaboration: Connect power system experts, workflow developers, and automation specialists
• Innovation: Create and share powerful workflows for various power system applications
• Education: Provide resources and examples for implementing workflow automation
• Standardization: Develop best practices for workflow design in the energy sector

Check out these demos showcasing PowerWorkflow in action:

• Grid Impact Evaluation for Large Load Interconnection: This demo showcases how Agentic AI can streamline the grid impact evaluation process for large loads like data centers. Using domain-adapted AI Agents equipped with tools such as PowerWorld, the workflow automates data analysis, contingency studies, and report generation—accelerating a process that traditionally requires extensive manual effort.

• Electricity Market Expert Demo: An automated workflow that handles electricity market inquiries, generates professional responses using an LLM, logs all actions for traceability, and supports human-in-the-loop review.

• FlowiseAI: Open-source visual workflow builder
• Dify: AI workflow development platform
• Coze: Bot development platform
• n8n: Workflow automation platform
• Gumloop: No-code workflow builder
• Relevance AI: No-code Agent and workflow builder
• Microsoft Copilot Studio: Enterprise workflow development

• LangGraph: Python framework for building agent workflows
• AutoGen: Microsoft's framework for multi-agent systems
• Crewai: Framework for collaborative AI agents

• A2A Protocol: An application level protocol that enables agents to collaborate in their natural modalities

We welcome contributions! Please see our Contributing Guidelines for details.

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

• Lin Dong, Qian Zhang

• All contributors who help make this project better
• The Power and AI Initiative (PAI) at Harvard SEAS