Scientific Agent Skills: A Library of Procedural Knowledge for Research Agents

NLP RAG Agent Tool Learning Autonomous Workflows AI4S Other AI4S
2100
2026年08月30日
当一个基于语言模型的智能体被要求分析某项实验时，它通常会返回可运行的代码；但该分析是否经得起推敲，则是另一个问题。一项经得起推敲的分析，依赖于一系列程序性决策：例如，学界公认应采用哪种统计检验方法、以哪个标识符命名空间为准、以及结果必须附带哪些必要的限定说明。为此，我们提出了“科学智能体技能库”（Scientific Agent Skills）——一个开源的程序集，涵盖16个实践领域（包括基因组学、化学信息学、医学影像、研究设计及科学传播等），共计163项此类标准化操作流程。每一项“技能”均以一个独立目录形式组织，其核心是一份经过版本控制、可供人类阅读的操作指南文件；智能体仅在执行相关任务时才加载该文件；该目录中通常还包含参考文献资料及可直接运行的脚本。本文未开展任务层面的效果评估，亦未报告宿主系统对各项技能的选择调用率。我们仅针对该文档语料库测度了两项指标：全部163项技能的常驻式描述文本共占用一个20万词元（token）上下文窗口的7.1%；而所有已文档化的典型工作流中，中位数长度占该窗口的23.9%；不过，在46个典型工作流中，若将每个所涉参考文件全部加载，则其中29个将超出窗口容量。本项目采用开放许可协议，源代码与文档可在 https://github.com/K-Dense-AI/scientific-agent-skills 公开获取。
A language-model agent asked to analyse an experiment will usually return working code. Whether the analysis is defensible is a different question. A defensible analysis depends on procedural choices: which test the field accepts, which identifier namespace is authoritative, and which caveats must accompany a result. We present Scientific Agent Skills, an open library of 163 such procedures in 16 areas of practice, including genomics, cheminformatics, medical imaging, study design and scientific communication. Each skill is a directory built around a versioned, human-readable instruction file. An agent loads the file only when a task calls for it; the directory often also contains reference material and runnable scripts. We report no task-level evaluation and no host selection rate. We measure two properties of the documentation corpus: the always-resident descriptions of all 163 skills cost 7.1% of a 200,000-token window, and the median documented workflow fits within 23.9% of it, although 29 of 46 would overflow if every reference file were loaded. Openly licensed and available at https://github.com/K-Dense-AI/scientific-agent-skills.
许愿