SkillsVote：智能体技能从收集、推荐到演化的生命周期治理

摘要

長程LLM代理會留下痕跡，這些痕跡可能轉化為可重複使用的經驗，但原始軌跡既嘈雜又難以掌控。我們將代理技能視為一種經驗模式，它將可執行腳本與關於程序流程的非可執行指導結合在一起。然而，開放技能生態系統包含冗餘、不均勻且對環境敏感的產物，而無差別的更新可能污染未來的上下文。我們提出SkillsVote，這是一個從收集、推薦到演化，針對代理技能生命週期的治理框架。SkillsVote對百萬規模的開源語料庫進行剖析，以評估環境需求、品質與可驗證性，接著為可驗證技能合成任務。在執行前，SkillsVote對結構化技能庫進行代理庫搜索，以揭示指導性的技能上下文。執行後，它將軌跡分解為與技能關聯的子任務，將結果歸因於技能使用、代理探索、環境與結果訊號，並且僅允許成功的可重用發現進入受證據控管的更新。在我們的評估中，離線演化使GPT-5.2在Terminal-Bench 2.0上提升高達7.9個百分點，而在線演化使SWE-Bench Pro提升高達2.6個百分點。總體而言，當系統控制曝光、信用與保存方式時，受治理的外部技能庫可以在無需模型更新的情況下，提升凍結代理的性能。

English

Long-horizon LLM agents leave traces that could become reusable experience, but raw trajectories are noisy and hard to govern. We treat Agent Skills as an experience schema that couples executable scripts, with non-executable guidance on procedures. Yet open skill ecosystems contain redundant, uneven, environment-sensitive artifacts, and indiscriminate updates can pollute future context. We present SkillsVote, a lifecycle-governance framework for Agent Skills from collection and recommendation to evolution. SkillsVote profiles a million-scale open-source corpus for environment requirements, quality, and verifiability, then synthesizes tasks for verifiable skills. Before execution, SkillsVote performs agentic library search over structured skill library to expose instructional skill context. After execution, it decomposes trajectories into skill-linked subtasks, attributes outcomes to skill use, agent exploration, environment, and result signals, and admits only successful reusable discoveries to evidence-gated updates. In our evaluation, offline evolution improves GPT-5.2 on Terminal-Bench 2.0 by up to 7.9 pp, while online evolution improves SWE-Bench Pro by up to 2.6 pp. Overall, governed external skill libraries can improve frozen agents without model updates when systems control exposure, credit, and preservation.