Vercel 評估指出：使用 AGENTS.md 定義 Agent 表現優於傳統的「技能 (Skills)」設定
Original: AGENTS.md outperforms skills in our agent evals
In its latest technical blog post, Vercel shared a significant finding regarding AI Agent architecture: in their Agent Evaluations (Agent…
Vercel 在最新的 Agent 評估中發現，使用 `AGENTS.md` 這種基於 Markdown 的宣告式文件來定義 AI Agent 的角色與指令，其成效顯著優於傳統為 Agent 綁定特定「技能（Skills/Tools）」的作法。這項發現將簡化開發者構建多 Agent 系統的流程，並提升 LLM 在複雜任務中的執行準確度與對齊表現，代表著 Agent 開發範式正朝向「文件驅動」轉移。
In its latest technical blog post, Vercel shared a significant finding regarding AI Agent architecture: in their Agent Evaluations (Agent Evals), using a declarative `AGENTS.md` Markdown file to define an Agent's behavior and architecture outperformed the traditional design pattern based on "Skills" or complex tool-calling.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Vercel Changelog →Summaries are AI-generated; the original article is authoritative.