Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add realchendahuang/dahuang-human-tone --skill dahuang-human-tonegit clone --depth 1 https://github.com/realchendahuang/dahuang-human-toneWrote 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.
[](https://agentmods.dev/skills/realchendahuang/dahuang-human-tone/dahuang-human-tone)<a href="https://agentmods.dev/skills/realchendahuang/dahuang-human-tone/dahuang-human-tone"><img src="https://agentmods.dev/badge/skills/realchendahuang/dahuang-human-tone/dahuang-human-tone/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.
<a href="https://agentmods.dev/skills/realchendahuang/dahuang-human-tone/dahuang-human-tone"><img src="https://agentmods.dev/badge/skills/realchendahuang/dahuang-human-tone/dahuang-human-tone.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00093 | $0.01915 |
| Opus 5 | $0.00046 | $0.00958 |
| Sonnet 5 | $0.00019 | $0.00383 |
| Haiku 4.5 | $0.00009 | $0.00192 |
Grade A, and why
dahuang-human-tone 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 13d 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.
How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dahuang Human Tone
把“模型在执行写作模板”的文本,改成“具体的人在具体场景里表达”。先保护内容,再改结构,最后恢复作者声音。
建立编辑合同
开始前确定五项:
- 模式:改写 / 诊断 / 局部处理。
- 强度:轻改 / 标准 / 深改。
- 体裁:技术、学术、社媒、营销、公开写作、聊天或叙事。
- 声纹:用户样本、原稿已有声音,或中性默认。
- 不可改变项:事实、术语、引用、格式、长度或指定段落。
能从请求和文本判断就直接执行。只有缺失信息会显著改变结果时,才问一个问题。
强度
| 强度 | 处理范围 |
|---|---|
| 轻改 | 只删最明显的模板壳和协作痕迹,不重排主体 |
| 标准 | 修结构、句式和节奏,保留原有论证顺序与声纹;默认 |
| 深改 | 把全文当作内容草稿重建表达,可重排段落,但仍不得改变内容账本 |
用户说“彻底”“大改”“重写”时用深改;技术、法律、学术方法段默认不超过标准,除非明确授权。
建立内容账本
改写前锁定:
- 事实、数字、日期、人名、链接、引语、引用与专有名词。
- 论点、结论、时间线、因果关系和不确定性。
- 代码、命令、接口、术语和行业含义。
- 作者的立场、幽默、偏见、犹豫、口头禅和有意修辞。
- 叙事中的人物声口、伏笔、钩子、情绪承接和情节功能。
保护含义和功能,不要求字面不动。深改也必须能逐项对回内容账本。
统一去味引擎
Pass 1:先修宏观结构
先找高杠杆问题:
- 文章是否被大纲骨架、同构段落或重复总结控制?
- 最重要的信息是否埋在背景、路标词和意义拔高后面?
- 多个弱信号是否其实落在同一个句子或段落?
合并重叠问题,不靠罗列数量制造严重感。优先修最影响读感的 1–3 处结构问题。
Pass 2:删壳并恢复具体性
- 删除不承载信息的开场、过渡、导读、总结和假互动。
- 拆掉高频二分对照、机械排比、三段式、权威揭示和金句公式。
- 把抽象宣传词换成原文已有的动作、对象、条件或结果;原文没有具体事实时直接删,不补新材料。
- 让真正的主语承担动作;被动语态在学术、法律或未知执行者场景中合理时保留。
- 相信读者能理解,不替每个比喻、情绪和结论再解释一遍。
Pass 3:修句子与连接
- 让句长、段长和标点服从内容,不人为制造“参差”。
- 删除同义循环、填充词、机械连接词和无源归因。
- 保留有功能的列表、破折号、粗体、被动语态和正式措辞。
- 处理假坦率开头、制造式短句、警句公式、标题后复述和“本次修改新增了……”式差异叙述。
详细模式只在需要时读取 references/patterns.md 与 references/deep-tells.md。
Pass 4:恢复作者声纹
优先级:
- 用户提供的写作样本。
- 原稿中已经存在、且与场景匹配的声音。
- 体裁默认。
按句长节奏、词汇级别、段落进入方式、转折习惯、标点、叙述距离和锋利度匹配声纹。没有样本时,不擅自添加第一人称、笑话、经历、俚语或情绪。
需要声纹校准时读取 references/voice-calibration.md;公开写作去壳后变得无菌时再读 references/soul.md。
Pass 5:验证与回滚
依次检查:
- 内容账本是否全部保留?
- 体裁规范是否仍然成立?
- 是否把作者磨成统一的“自然口吻”?
- 是否新增原文没有的事实、来源、经历或情绪?
- 大声读或默读时,是否仍有新闻稿、讲义或模板脚本感?
做一次“为什么它还像 AI”的反向检查,只修新发现的高杠杆问题。第二轮没有实质改善就停止,不无限迭代。
如果原文已经自然且符合场景,允许不改或只改一两处。去味不是必须留下大量 diff。
体裁优先
体裁规则优先于通用“人味”:
- 技术、学术、中英混排、营销、社媒、聊天和叙事的具体边界见
references/genre-routing.md。 - 高误伤风险、引用、术语和正式表达见
references/guardrails.md。 - 长文只有出现明显同构或节奏重复时才读
references/shape-audit.md。
输出合同
默认改写
只输出改写后的正文,不加“优化如下”、评分或自我表扬。
诊断
## 诊断
### 最优先
- **原文证据**:{短引用}
**问题**:{合并后的结构或表达问题}
**影响**:{为什么不适合当前体裁}
**建议**:{最小修改方向}
### 可保留
- {容易被误伤、但符合作者或体裁的写法}
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 346 B
- examples/audit.md 2.9 KB
- examples/before-after.md 6.0 KB
- examples/regression-cases.md 4.4 KB
- references/deep-tells.md 16 KB
- references/genre-routing.md 5.1 KB
- references/guardrails.md 9.3 KB
- references/patterns.md 23 KB
- references/shape-audit.md 3.2 KB
- references/soul.md 4.3 KB
- references/sources.md 1.8 KB
- references/voice-calibration.md 3.7 KB
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.
- 13d ago First seen · 161 lines · 93 tokens per session scan A 79e7c6a1fff0
dahuang-human-tone is a skill published in the GitHub repository realchendahuang/dahuang-human-tone (11 stars, last pushed 2mo ago), licensed MIT. It adds 93 tokens to every session and 1,915 once invoked, about $0.0005 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…