Microsoft's SkillOpt shows optimized agent skill artifacts transfer across model scales and between Codex and Claude Code
Original title:Microsoft’s SkillOpt Shows Optimized Agent Skill Artifacts Transfer Across Model Scales and Between Codex and Claude Code Harnesses
AI summary
SkillOpt, proposed by Microsoft together with teams from Shanghai Jiao Tong University, Tongji University and Fudan University, trains a single skill document through text-space optimization while freezing the target model, allowing the optimized skill artifact to transfer across model scales and toolchains. A SpreadsheetBench skill optimized on Codex scored 81.8 after deployment to Claude Code, exceeding the 80.4 achieved by the latter training its own skill. All 4 cross-model, 4 cross-toolchain and 3 cross-benchmark transfer results were higher than the target's skill-free baseline.
Key Takeaways
- SkillOpt,proposed by Microsoft together with teams from Shanghai Jiao Tong University,Tongji…
- A SpreadsheetBench skill optimized on Codex scored 81.
- 8 after deployment to Claude Code,exceeding the 80.
- 4 achieved by the latter training its own skill.
- All 4 cross-model,4 cross-toolchain and 3 cross-benchmark transfer results were higher than the…
Quick Facts
- Published
- Updated
- Source
- MarkTechPost(RSS)
- Category
- Papers & Research
- AI score
- 70
跨 harness 迁移实验把 agent 技能从单环境优化推进到可复用文本技能文件,Codex 练出的技能在 Claude Code 上超过域内训练结果,对维持多工具一致行为提供直接实证。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Microsoft's SkillOpt shows optimized agent skill artifacts transfer across model scales and between Codex and Claude Code. August 6, 2026. https://new.nat6.net/en/news/cmsgsgz530fluro5qu0vnw8s0/