SkillOpt: Executive Strategy for Self-Evolving Agent Skills

Adapting language agents often means teaching them domain procedures: where to look, which tools to call, how to verify intermediate results, and how to format outputs for a grader. Existing context-level adaptation usually either hand-writes such procedures, generates them once, or lets skill artifacts grow through loosely controlled self-revision. We ask whether a skill can instead be trained as the compact external state of a frozen agent. SkillOpt is a harness-agnostic text-space optimizer for a single natural-language skill document. It runs the frozen target model on scored rollout batches, uses a separate optimizer model to reflect over success and failure minibatches, proposes structured add/delete/replace edits, ranks them under a textual learning-rate budget, and accepts a candidate only if it improves held-out selection performance. Rejected-update memory, slow cross-epoch guidance, and optimizer-side meta skill make the editing loop behave like bounded training while leaving deployment as a single best_skill.md. Across six direct-chat benchmarks covering QA, spreadsheets, documents, math, and embodied decision making, improves GPT–5.5 over no skill by 21.5 points on average, with positive gains on every task and the best measured result on five of six. Codex- and Claude-Code-style harness runs show that the same artifact remains useful in tool-backed execution. Ablations and transfer studies identify validation-gated, bounded edits and update memory as key to stable, portable skill learning.