elvis@omarsar0
37AI 编辑部评分,满分 100

CLAUDE.md 为何不断膨胀?研究揭示修复方案

2026-08-13 02:20· 50分钟前
AI 导读

一项针对 1,867 个仓库中 247,694 条指令生命周期的研究发现,CLAUDE.md 等文件会无限膨胀:提示词在其生命周期内增长超三倍,每次提交净增 4.9 条指令,且指令越旧越难被删除。研究提出的修复方案是为指令添加注释说明理由,在可验证环境中可消除 99.3% 的多余指令,并将真实智能体指令遵循率提升最高 23.1%。

Why does CLAUDE.MD keep growing?

If you maintain a CLAUDE.md or an AGENTS.md, this one is worth your time.

(bookmark it)

This work traces why these files grow without bound. Appending an instruction is free. Deleting one after its rationale is gone costs exponential verification, so nobody removes anything.

Across 247,694 instruction lifetimes in 1,867 repositories, prompts more than tripled over their lifetime and gained 4.9 net instructions per commit. The older an instruction gets, the less likely anyone is to delete it.

The proposed fix is comments. Writing down the reasoning behind an instruction removed 99.3% of excess instructions in verifiable settings and improved real agentic instruction-following by up to 23.1%.

Paper: https://arxiv.org/abs/2608.11095

Track more trending AI papers in our academy: https://academy.dair.ai/

来源:elvis · x.com

CLAUDE.md 为何不断膨胀?研究揭示修复方案

elvis · @omarsar0 · X·2026-08-13 02:20·50分钟前
AI 导读

一项针对 1,867 个仓库中 247,694 条指令生命周期的研究发现,CLAUDE.md 等文件会无限膨胀:提示词在其生命周期内增长超三倍,每次提交净增 4.9 条指令,且指令越旧越难被删除。研究提出的修复方案是为指令添加注释说明理由,在可验证环境中可消除 99.3% 的多余指令,并将真实智能体指令遵循率提升最高 23.1%。

Why does CLAUDE.MD keep growing?

If you maintain a CLAUDE.md or an AGENTS.md, this one is worth your time.

(bookmark it)

This work traces why these files grow without bound. Appending an instruction is free. Deleting one after its rationale is gone costs exponential verification, so nobody removes anything.

Across 247,694 instruction lifetimes in 1,867 repositories, prompts more than tripled over their lifetime and gained 4.9 net instructions per commit. The older an instruction gets, the less likely anyone is to delete it.

The proposed fix is comments. Writing down the reasoning behind an instruction removed 99.3% of excess instructions in verifiable settings and improved real agentic instruction-following by up to 23.1%.

Paper: https://arxiv.org/abs/2608.11095

Track more trending AI papers in our academy: https://academy.dair.ai/

来源:elvis· x.com