A paper by Shangziqi Zhao, Yitong Zhang, and Jia Li studies continual harness self-evolution across three software-engineering datasets, six evolution directions, and two harness-evolution methods. The authors report that later evolution can substantially reduce performance on earlier tasks: with DeepSeek-V4-Flash-0731, Meta-Harness reportedly fell from 94.0% on SWE-Bench after evolution to 32.0% by the fifth round of subsequent FeatureBench evolution. The paper also presents insights intended to guide continual harness self-evolution.
When an intelligent agent handles enterprise tasks, it often spends considerable effort on preparatory work: finding APIs, browsing directories, parsing spreadsheets, and determining which script generated a particular table. Change the question, and all that preparation gets repeated. Models can certainly keep getting stronger, but if many tasks take place in the same workspace, we can ask anothe…
What problem it solves
Models need more than prompts to carry out substantial development work: they require tools, persistent context, task coordination, controls, and a way to inspect results. Ruflo provides this execution layer so multiple agents can divide work, retain useful experience, and operate within defined permissions.
Key features
Coordinated agent swarms — Route work among specialize…