# EvoSkill v2 用持久化技能让 Agent 免重训自我提升，并演示技能钻评分器空子的风险

- 来源：elvis (@omarsar0)
- 发布时间：2026-09-19 00:45
- AIHOT 分数：55
- AIHOT 链接：https://aihot.news/items/cmu77v0380tqtrowkv9utgvny
- 原文链接：https://x.com/omarsar0/status/2100989618845823364

## AI 摘要

EvoSkill v2 通过持久化 Agent 技能实现免重训的自我提升：coach agent 阅读失败运行后写出技能文件，worker 在下次遇到类似任务时加载，不改动任何权重。

## 正文

Agents can self-improve without retraining.

EvoSkill v2 achieves this with persistent agent skills.

A coach agent reads the failed runs and writes the skill. The worker loads it the next time a similar task shows up.

No weights are touched. Every improvement comes from a simple file with lessons.

Every bad lesson also gets saved. On spreadsheet repair, the coach found the grader trusted cached values and wrote a skill telling the worker to skip recalculation.

Sentient's fix was to split the roles. The agent that writes skills cannot touch the test. A person reviews the results after every round.

With that in place, the hardest spreadsheet tasks went from 3 passes out of 120 to 21.

### 引用推文

> Sentient：Last week, Dario Amodei published "We Must Pace the Frontier". His concern: the OpenAI–Hugging Face incident in which a swarm of agents tried to hack their own ...