# Harness Handbook：让智能体框架代码可读、可导航、可编辑

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-07-24 14:48
- AIHOT 分数：37
- AIHOT 链接：https://aihot.virxact.com/items/cmryl335a02c8rolggcuyov3r
- 原文链接：https://x.com/rohanpaul_ai/status/2080545686919987619

## AI 摘要

Harness Handbook 将代码仓库按运行时行为重新组织，并链接已验证的源码位置，帮助编码智能体找到更多需修改的代码位置，同时减少规划 token 使用量。在 Codex 和 Terminus-2 请求上，该方法将计划成功率分别提升 10.0 和 18.9 个百分点，同时 token 消耗降低 12.7% 和 8.6%。

## 正文

Beautiful paper on agent harness

Agent harnesses fail to evolve cleanly because developers often know what must change， but not where the behavior lives.

A behavior-based code handbook helped coding agents find more required edit sites while using fewer planning tokens.

Agent harnesses control prompts， tools， memory， and execution， but one behavior often spans distant files， functions， and states.

This makes code changes risky because a request describes what should change， while repositories reveal only where code lives.

Harness Handbook reorganizes a repository around runtime behaviors， then links each behavior to verified source locations.

It combines static analysis with LLM descriptions， 3 detail levels， shared-state maps， and automatic updates after edits.

Its guided search starts broad， narrows to likely code， follows call and state links， then checks live source.

Across Codex and Terminus-2 requests， it raised plan win rates by 10.0 and 18.9 points while cutting token use by 12.7% and 8.6%.

- arxiv. org/abs/2607.13285

Title： "Harness Handbook： Making Evolving Agent Harnesses Readable，Navigable， and Editable"