Anthropic describes automating eval design and hillclimbing with Claude Code and the claude-api skill
Original title:Automating eval design and hillclimbing with Claude
AI summary
Anthropic added two commands to the claude-api skill, /claude-api build-eval and /claude-api hillclimb. The former builds an eval set within a codebase through an interview-style process and validates the grader, while the latter iteratively improves an application with one change at a time, train/test separation and overfitting detection.
Key Takeaways
- Anthropic added two commands to the claude-api skill,/claude-api build-eval and /claude-api…
- The former builds an eval set within a codebase through an interview-style process and validates…
- train/test separation and overfitting detection.
Quick Facts
- Published
- Updated
- Category
- Tutorials & Tips
- Topic
- Deployment Engineering
- AI score
- 67
Anthropic 官方讲解了评测设计的关键原则,并给出 build-eval 和 hillclimb 两个命令的用法与真实优化案例,可迁移到自己的应用调优。
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: Anthropic describes automating eval design and hillclimbing with Claude Code and the claude-api skill. September 28, 2026. https://new.nat6.net/en/news/f846kwxx0jt5sqhzifx3zdfrb/