# Prime Agent 技术报告：记忆层级机制解析

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-08-26 06:39
- AIHOT 分数：36
- AIHOT 链接：https://aihot.virxact.com/items/cmt99ajk80dn5rolyo7q1wg3s
- 原文链接：https://x.com/rohanpaul_ai/status/2092381340968910922

## AI 摘要

Prime Agent 发布技术报告，该 harness 曾助 Opus 5 在 ARC-AGI-3 达 95.5%。其核心机制为记忆层级：模型权重与活动上下文之下，设持久 IPython 会话及磁盘存储的历史、技能与提示词库，模型通过代码在各层级间移动状态。长输入作为 REPL 变量留存，供智能体搜索与转换，将长上下文工作转为信息管理问题。

## 正文

Prime Agent, the harness that put Opus 5 at 95.5% on ARC-AGI-3, now has a technical report.

The mechanism is a memory hierarchy. Model weights and active context sit underneath a persistent IPython session and a disk-backed store of histories, skills and prompts, and the model moves state between those levels with code instead of having it compacted away.

Long inputs stay in the REPL as variables the agent can search and transform, so long-context work becomes an information-management problem rather than a reading problem.

– arxiv. org/abs/2608.23552

Title: "Prime Agent: A Self-Improving RLM Harness"