Microsoft's SkillOpt Optimizes AI Agent Skills

Microsoft Research has unveiled SkillOpt, a new method for optimizing AI agent skills without changing model weights. SkillOpt treats skills as trainable parameters, turning skill editing into a controlled optimization process. This approach has shown consistent performance improvements across multiple benchmarks and models, indicating that optimized skills capture reusable knowledge. The ability to transfer these skills across different models and tasks suggests a significant advancement in creating adaptable AI agents.