This repository hosts the WaveForge methodology, a highly structured software development workflow designed for seamless collaboration between human and AI developers.
Our vision is to democratize software creation. We aim to provide anyone with an idea their own personal AI agent development team. By engaging in "Document-Oriented Programming"—simply articulating needs and goals through a series of structured documents—users can bring their vision to life, ensuring the final product aligns perfectly with their original intent. This methodology is the first step toward that future.
- Structured: All development activities follow predefined processes and templates, with clear guidance at every step.
- Transparent: The AI's working status and plans are visible in real-time, allowing human collaborators to understand its progress at any time.
- Collaborative: Clear human-AI interaction "gates" ensure a perfect blend of AI autonomy and human oversight.
- Document-Driven: The development process itself produces high-quality, interconnected documentation, forming a valuable knowledge base.
- Self-Improving: A built-in reflection and memory update mechanism enables this workflow to learn and continuously improve.
While the principles of WaveForge are universal, this methodology is specifically engineered for and deeply integrated with the Cursor AI code editor. The .cursor-en/ and .cursor-zh/ directories contain configurations that directly leverage Cursor's advanced features. By providing this structured framework, we transform Cursor into a specialized, project-aware development partner, enhancing its capabilities to perform complex tasks with unparalleled consistency.
We have two versions of the documentation, one in English and one in Chinese. Please refer to the directories below for detailed documentation, including roles, templates, rules, and example workflows with Cursor.
This project is licensed under the MIT License. See the LICENSE file for details.