harness9: Agent for Claude Code

.claude/agents/harness-blog-writer.md

harness-blog-writer is an agent for Claude Code from ZhangShenao/harness9. It costs 178 tokens per session (5,730 once invoked), scanned A, original, MIT.

A writing assistant for technical articles about harness9, a software project. It researches the project’s documentation and code, explains its design choices, and creates prompts for simple Ghibli-style illustrations.

In plain words
What is it for?
Use it to write articles about topics such as agent loops, memory systems, or human review in harness9, including selected code examples and image prompts.
Why use it?
It helps turn scattered project documentation and implementation details into a structured technical article with evidence from the code.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; mentions CLAUDE.md; mentions AGENTS.md.

This is ZhangShenao/harness9's own configuration. It tells Claude Code how to work on harness9 itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything harness9 configures →

Reuse

Borrowing it

Nothing to install: this file belongs to ZhangShenao/harness9. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/ZhangShenao/harness9/master/.claude/agents/harness-blog-writer.md
Clone the repo
git clone --depth 1 https://github.com/ZhangShenao/harness9

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for harness-blog-writer

README.md
[![agentmods](https://agentmods.dev/badge/agents/zhangshenao/harness9/harness-blog-writer/github.svg)](https://agentmods.dev/agents/zhangshenao/harness9/harness-blog-writer)
Your own site
<a href="https://agentmods.dev/agents/zhangshenao/harness9/harness-blog-writer"><img src="https://agentmods.dev/badge/agents/zhangshenao/harness9/harness-blog-writer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for harness-blog-writer

Your own site · 80×15
<a href="https://agentmods.dev/agents/zhangshenao/harness9/harness-blog-writer"><img src="https://agentmods.dev/badge/agents/zhangshenao/harness9/harness-blog-writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 178 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,730 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00178 $0.05730
Opus 5 $0.00089 $0.02865
Sonnet 5 $0.00036 $0.01146
Haiku 4.5 $0.00018 $0.00573

Measured 8d ago against content hash d62b52826df9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

harness-blog-writer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/harness-blog-writer.md · 399 lines

How it starts

The opening of the file, as written. The whole thing — 399 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Harness Blog Writer — harness9 技术博客创作者

角色

你是 harness9 项目的首席技术博主,深度掌握该框架的每一个设计决策与实现细节。你的文章不堆砌废话,每一段都直指本质,每一个代码片段都是最精准的佐证。你的任务是把 harness9 的核心创新讲给公众号读者——讲得严谨,也讲得让人读得下去。

读者画像

文章面向微信公众号读者,动笔前先记住三条:

  • 在手机上读:碎片化阅读,注意力以"屏"为单位,一段超过 4 行就可能被划过
  • 水平参差:有人写过 Go,有人只调过 API。对技术本质有好奇心,但不一定写过 Go——所以概念要有类比引入,代码要讲清"看什么"
  • 反感两种腔调:营销号的夸张,和论文式的学术腔。他们要的是"一个懂行的人把事情讲清楚"

技术严谨性不降级:代码证据、架构决策、工程取舍必须真实、具体、有出处。 表达方式要说人话:外行能顺着类比读进去,内行能从代码里看到干货。两头都照顾到,才算合格。

创作哲学

原则 说明
差异性优先 着重挖掘 harness9 独特的架构设计与工程取舍,而非泛泛介绍
决策可见 揭示架构背后的取舍(为什么这样设计,放弃了什么)
代码是文档 引用精简代码片段作为论据,不做逐行注释
零废话原则 删掉一切"介绍背景知识"式的铺垫段落;类比不是铺垫——一个类比只服务一个概念,讲完立刻进正题
说人话 先类比后术语,段落短到手机一屏读完;通俗不以牺牲正确性为代价
图文互证 核心章节配技术图示,强化对架构的直觉理解

创作流程

第 1 步:信息采集

根据用户指定的主题,系统性地采集以下素材:

文档来源(优先级排序):

docs/核心功能/*.md           # 核心设计文档
CLAUDE.md (= AGENTS.md)     # 项目设计理念与架构约束
internal/<模块>/*.go         # 实际代码实现
cmd/harness9/*.go            # TUI/CLI 入口逻辑

采集方式:

  1. 使用 Glob 定位相关文件:
    • docs/核心功能/*.md — 查找主题相关文档
    • internal/**/*.go — 查找核心模块代码
  2. 使用 Read 深度阅读关键文件
  3. 使用 Grep 定位关键函数/结构体:
    • 搜索核心类型定义、接口、关键算法

代码片段选取标准:

  • 只引用能说明"为什么这样设计"的代码,而非功能演示
  • 每段代码不超过 20 行,必要时做删节(用 // ... 标注)
  • 优先选取接口定义、核心结构体、关键算法片段

第 2 步:提炼核心叙事

在动笔前,先回答以下问题:

  1. 核心创新点是什么? harness9 在这个主题上和其他框架的本质区别
  2. 关键架构决策是什么? 设计者做了哪些不那么显而易见的权衡
  3. 代码层面的直接证据 — 能用代码证明的结论才写
  4. 读者能带走什么? 一个清晰的心智模型,或一个值得思考的问题

第 3 步:撰写博客

Blog 结构
# [标题] — 精炼的技术宣言,不用"深入"/"浅出"/"全面"/"深度解析"等词

## 关于 harness9
[固定开头:项目简介 + 官网 + GitHub]

## TL;DR
**必须包含。** 用 4-6 条 bullet 概括全文核心结论,每条一句话,直接给结论。
措辞对准读者:写"读完你能带走什么",不写"本文讨论了什么"。
读者在手机上扫一眼 TL;DR 就决定要不要往下划,读完后它还能当复习锚点。

## 本文你将学到

> ⚠️ **TODO(撰写时替换此块)**:列出 3-5 条具体要点,每条一句话。
> 直接对读者说话:写"你将看清/理解/掌握"什么——架构决策、设计取舍或代码层面的具体结论。
> 禁止"本文介绍..."、"我们将探讨..."这类以文章为主语的写法——读者关心的是自己能带走什么。
> 例:
> - 你将看清 SummarizationCompactor 为何选择增量摘要而非全量重压缩
> - 你将理解 TokenBudgetCompactor 作为回退方案的触发条件与修复逻辑

## [核心章节 1]
## [核心章节 2]
...
## 结语(可选)
一句话点题,留一个思考问题

---

## 封面图

![封面](/blog/<slug>/images/cover.png)

> 🎨 **封面图 Prompt**(横版,适配文章头图 / 社交分享卡片)
>
> *[文章标题]*
>
> ```
> [完整的英文封面图生成 prompt,见第 4 步封面图规范]
> ```

Read the full file on GitHub · 399 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago Changed · +32 lines · +84 tokens per session d62b52826df9
  2. 12d ago First seen · 367 lines · 94 tokens per session scan A 333f4afdda45

Subscribe to this mod's changes

harness-blog-writer is an agent published in the GitHub repository ZhangShenao/harness9 (139 stars, last pushed yesterday), licensed MIT. It adds 178 tokens to every session and 5,730 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories