Subagents vs Agent Skills: Executing Reusable Knowledge for Long-Horizon Agentic Tasks

子智能体与智能体技能：长程智能体任务中可复用知识的执行策略

Abstract: How can language model agents effectively leverage libraries of reusable knowledge to solve long-horizon tasks? Recent work has increasingly focused on agent skills: reusable capabilities represented as skill packages, i.e., multi-file bundles containing instructions, scripts, and other resources that help agents perform specific tasks.

摘要： 语言模型智能体如何有效地利用可复用知识库来解决长程任务？近期的研究日益关注“智能体技能”（agent skills）：即以技能包形式呈现的可复用能力，例如包含指令、脚本及其他资源的复合文件包，旨在帮助智能体执行特定任务。

Agent skills are typically executed by loading their skill instructions into an agent’s context and relying on the agent to follow them. As task horizons grow, however, this approach becomes increasingly brittle, because reasoning quality degrades as more information accumulates in the context window.

智能体技能通常通过将技能指令加载到智能体的上下文中，并依赖智能体遵循这些指令来执行。然而，随着任务跨度的增加，这种方法变得愈发脆弱，因为随着上下文窗口中积累的信息增多，推理质量会随之下降。

We investigate an alternative approach in which skill packages are instead invoked as subagents. Rather than loading skill instructions into the main context, subagent execution spawns fresh context windows dedicated to solving individual subtasks.

我们研究了一种替代方案，即将技能包作为“子智能体”（subagents）来调用。子智能体执行方式不再将技能指令加载到主上下文中，而是为解决各个子任务创建全新的上下文窗口。

We show that subagent execution outperforms agent-skill execution when skill packages expose clear input-output contracts and their instructions encode the procedural knowledge needed to fulfill those contracts. The tradeoff is additional communication overhead, as extra tokens are required to coordinate between the main agent and its subagents.

研究表明，当技能包提供清晰的输入输出契约，且其指令编码了实现这些契约所需的程序性知识时，子智能体的执行效果优于智能体技能的执行效果。其代价是额外的通信开销，因为主智能体与子智能体之间需要额外的 Token 来进行协调。

Our results show that the benefit of reusable knowledge depends not only on its content, but also on how it is organized and invoked.

我们的研究结果表明，可复用知识的价值不仅取决于其内容，还取决于其组织和调用的方式。