让网页智能体的技能可修复、可验证，提升稳定性与复用性。
ContractSkill: Repairable Contract-Based Skills for Multimodal Web Agents
- 将技能转为带显式流程的可执行代码，支持精准检查和局部修复
- 在VisualWebArena上性能提升，修复后仍能独立使用
- 适合需要稳定、可维护智能体系统的研发团队
网页智能体自动生成的技能常不稳定，甚至降低性能。我们指出问题不仅在于生成质量，更在于技能隐式存在，无法验证或局部修复。为此提出ContractSkill框架，将草稿技能转化为具有显式程序结构的可执行实体，实现确定性验证、故障定位和最小化本地修复。这使技能优化从全量重写变为单个技能的局部编辑。在VisualWebArena的真实环境实验中，ContractSkill表现有效；而在MiniWoB的受控测试中，验证了机制有效性。在匹配迁移层条件下，修复后的技能即使移除源模型仍可复用，表明其在同类别基准间具备可移植性，而非跨基准泛化。结果表明，核心挑战并非生成技能本身，而是使其显式、可执行且可修复。代码已开源：https://github.com/underfitting-lu/contractskill.git。
原文摘要 · Abstract (English)
Self-generated skills for web agents are often unstable and can even hurt performance relative to direct acting. We argue that the key bottleneck is not only skill generation quality, but the fact that web skills remain implicit and therefore cannot be checked or locally repaired. To address this, we present ContractSkill, a framework that converts a draft skill into an executable artifact with explicit procedural structure, enabling deterministic verifica tion, fault localization, and minimal local repair. This turns skill refinement from full rewriting into localized editing of a single skill artifact. Experiments on VisualWebArena show that Contract Skill is effective in realistic web environments, while MiniWoB provides a controlled test of the mechanism behind the gain. Under matched transfer layers, repaired artifacts also remain reusable after removing the source model from the loop, providing evi dence of portability within the same benchmark family rather than full-benchmark generalization. These results suggest that the central challenge is not merely generating skills, but mak ing them explicit, executable, and repairable. Code is available at https://github.com/underfitting-lu/contractskill.git.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。