SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.
Framework-agnostic implementation of ReasoningBank (Ouyang et al., ICLR 2026): agents that learn from successful and failed trajectories, with memory-aware test-time scaling (MaTTS).
