This category supports the Agentic Engineering Working Group, a community effort to develop open, reusable infrastructure for AI-assisted computational modeling.
Our focus is not on discussing the latest AI models or products. Instead, we aim to capture computational modeling expertise as open guidance, reusable workflows, and agent skills that help researchers apply established methods more consistently and transparently.
We welcome discussion of topics such as:
- Agent skills for computational modeling and research software
- Methodological guidance and scientific workflows
- Documentation, metadata, and FAIR practices
- Reproducibility, verification, and evaluation
- HPC, automation, and research infrastructure
- Demonstrations, prototypes, and community use cases
- Reviews and feedback on proposed skills
This category complements existing discussions about computational models and research software by focusing specifically on the design, evaluation, and governance of reusable agent skills and AI-assisted workflows.
The accompanying comses/skills repository serves as the primary location for specifications, implementations, and documentation. This forum is the place to discuss ideas, propose new skills, coordinate development, review designs, share experiences, and help establish community best practices.
Whether you’re actively using AI agents, curious about their potential, or interested in improving computational modeling workflows, we encourage you to participate.