SkillX:面向智能体的技能知识库自动化构建系统
SkillX: Automatically Constructing Skill Knowledge Bases for Agents
April 6, 2026
作者: Chenxi Wang, Zhuoyun Yu, Xin Xie, Wuguannan Yao, Runnan Fang, Shuofei Qiao, Kexin Cao, Guozhou Zheng, Xiang Qi, Peng Zhang, Shumin Deng
cs.AI
摘要
基于经验学习对于构建强大语言模型智能体至关重要,然而当前主流自进化范式效率低下:智能体在孤立环境中学习,从有限经验中反复发现相似行为,导致重复探索与泛化能力不足。为此我们提出SkillX——一个全自动构建即插即用技能知识库的框架,该知识库可跨智能体与环境复用。SkillX通过三项协同创新技术实现全自动流水线:(1)多层级技能设计,将原始轨迹提炼为战略规划、功能技能与原子技能的三层架构;(2)迭代式技能优化,根据执行反馈自动修订技能以持续提升知识库质量;(3)探索式技能扩展,主动生成并验证新技能以突破初始训练数据的覆盖范围。基于强骨干智能体(GLM-4.6),我们自动构建可复用技能库,并在AppWorld、BFCL-v3和τ²-Bench等具挑战性的长周期人机交互基准测试中评估其迁移性。实验表明,当SkillKB接入较弱基座智能体时,能持续提升任务成功率与执行效率,印证了结构化分层经验表征对通用智能体学习的重要性。代码即将发布于https://github.com/zjunlp/SkillX。
English
Learning from experience is critical for building capable large language model (LLM) agents, yet prevailing self-evolving paradigms remain inefficient: agents learn in isolation, repeatedly rediscover similar behaviors from limited experience, resulting in redundant exploration and poor generalization. To address this problem, we propose SkillX, a fully automated framework for constructing a plug-and-play skill knowledge base that can be reused across agents and environments. SkillX operates through a fully automated pipeline built on three synergistic innovations: (i) Multi-Level Skills Design, which distills raw trajectories into three-tiered hierarchy of strategic plans, functional skills, and atomic skills; (ii) Iterative Skills Refinement, which automatically revises skills based on execution feedback to continuously improve library quality; and (iii) Exploratory Skills Expansion, which proactively generates and validates novel skills to expand coverage beyond seed training data. Using a strong backbone agent (GLM-4.6), we automatically build a reusable skill library and evaluate its transferability on challenging long-horizon, user-interactive benchmarks, including AppWorld, BFCL-v3, and τ^2-Bench. Experiments show that SkillKB consistently improves task success and execution efficiency when plugged into weaker base agents, highlighting the importance of structured, hierarchical experience representations for generalizable agent learning. Our code will be publicly available soon at https://github.com/zjunlp/SkillX.