arXiv：cs.AI· Samuele Giampieri·· 22 小时前
Cortex：用测量门控对编程智能体的技能与规则进行演化与剪枝
Cortex: Measurement-Gated Evolution and Pruning of Coding-Agent Skills and Rules
AI 导读
Cortex 是一个针对编程智能体 harness（项目文件、技能与规则）的测试式演化工具，把开发者会话中的修正保存为可重放任务，逐条写入技能或规则，仅当重复运行显示净收益且无任务明显变差时才保留。
正文
Abstract:Coding agents such as Claude Code also follow text files that developers write for them: a project file, skills the agent may choose to open, and rules that load automatically when the agent reads matching files. We call these files the agent's harness. Developers add to it whenever they correct the agent, but one success proves little: the same agent can pass a task once and fail it the next time.
We present Cortex, a tool that tests changes to the harness instead of trusting them. It saves fixes from the developer's sessions as replayable tasks, finds the corrections that recur, writes one skill or rule at a time, and keeps it only if repeated runs with and without it show a net gain and no task made noticeably worse. Existing text is tested the same way before it is deleted, and Cortex neither adds nor deletes text the agent never used.
We tested Cortex with Claude Haiku 4.5 on a repository built for the purpose, with conventions the model tends to break, one area with nothing to learn, and a script playing the developer. On held-out tasks, the harnesses it built made the agent follow the conventions far more often than no skills or rules did, left the comparison area unchanged, and added no text read on every turn. Rules, which load automatically, did most of the work; skills helped only when offered with the files a task touches, and those offered on every turn were rarely opened. Cortex kept none of the useless or harmful suggestions we planted, but the agent never opened the useless ones or the harmful skills, and it reached no verdict on the harmful rules, so the test shows that Cortex keeps no unused text, not that it detects damage. Under the stronger Claude Sonnet 5 a single run's harness added little, and on a real open-source project, where the agent was rarely corrected, Cortex found nothing to learn. We release the tool, the lab and its data.
| Comments: | 46 pages, 22 figures, 8 tables. Abstract abridged. Code and data: this https URL |
| Subjects: | Software Engineering (cs.SE); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.14216 [cs.SE] |
| (or arXiv:2610.14216v1 [cs.SE] for this version) | |
| https://doi.org/10.48550/arXiv.2610.14216 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Samuele Giampieri [view email]
[v1]
Fri, 9 Oct 2026 18:29:48 UTC (2,921 KB)
来源：arXiv：cs.AI · arxiv.org