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 Job-Yang/jobbyang-ai-skills --skill haohao-shuohuagit clone --depth 1 https://github.com/Job-Yang/jobbyang-ai-skillsWrote 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/job-yang/jobbyang-ai-skills/haohao-shuohua)<a href="https://agentmods.dev/skills/job-yang/jobbyang-ai-skills/haohao-shuohua"><img src="https://agentmods.dev/badge/skills/job-yang/jobbyang-ai-skills/haohao-shuohua/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/job-yang/jobbyang-ai-skills/haohao-shuohua"><img src="https://agentmods.dev/badge/skills/job-yang/jobbyang-ai-skills/haohao-shuohua.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00225 | $0.10685 |
| Opus 5 | $0.00112 | $0.05343 |
| Sonnet 5 | $0.00045 | $0.02137 |
| Haiku 4.5 | $0.00022 | $0.01069 |
Grade A, and why
haohao-shuohua 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 12d 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 — 291 lines — stays where its author put it; the contents beside it link to each section on GitHub.
好好说话 · 中文写作底层清洗
这是一套常驻的写作规范。 只要产出中文,就默认从这里过一次,让文字回到「像人说的中文」。
一句话总纲:保事实不动,去 AI 味,加中文味,不许造词充深刻。
尺度跟着任务走,从含义出发(这是本技能的根,先看这条):让你改一篇文章,整篇就是上下文,做文章级清洗,别把文章拆成一句句挨个翻译。让你改一段,就照这段的整体含义改,别盯着段里孤零零一句。动手前先读懂全文的结构和含义、每段在讲什么,再从含义出发整体改写,最后才拆到每句该说什么、怎么改。段落和句子的设计本身就可能带 AI 味,只动句子、不碰整段和全篇,永远改不好。站在全篇看,有的句子根本不用留,有的整段都可以砍掉——这种删减只有文章级视角才看得见。靠规则套公式永远套不好,一定从含义出发;规则是兜底的,不是改写的起点。
手别太痒(跟上一条不冲突):尺度放到全篇,不等于把每句都重写一遍。没有 AI 味、意思也清楚的句子,别为了"改得更好"去动它。每改一处先问「这处原文有毛病吗?」没有就放回去。"没忍住把好句子也重写了"是本技能第一号翻车原因。 一句话:看得宽(全篇兜底),动得准(只动该动的)。
跟三思而后行的分工:整篇骨架该怎么搭、改一段会不会伤到全文结构,那是
三思而后行技能的事。好好说话管的是——在给定的任务尺度里,怎么把话写好、改好。
随手回消息、口语转一段这类不写回持久面的,看这一页够了。写回持久面的成稿(手记、硬文、对外文档)的完整流程、加码/收手怎么调、修改报告模板在
references/workflow.md。
使用面(先看这里)
- 默认常驻:任何写中文的场景,模型都应该把本技能的原则和红线内化到起草过程里,不要等用户喊「改稿」再启动。用户说「写篇文档」「起草个方案」「帮我写段周报」,就已经命中本技能。
- 显式触发:用户主动喊「过一遍好好说话」「改成中文的样子」「这段像 AI」,或直接让你改写/润色一篇东西——都按默认开大档洗,跑完整流程。
- 被上层调用:任何上层写作技能(想法硬文、畅想、技术科普等)在自己的「去 AI 味/收尾清洗」环节,都应该把成稿交本技能开大档过一次,省得各自维护一份会走样的 AI 味清单。依赖是单向的,上层调用它,它不反过来依赖上层。
自我豁免:说明书的条目免检,说明书里的话要过
本技能自己就是规范文本,规范文本里有大量编号、量词标题、并列条目(「四条原则」「四条零容忍」「第五层」)。这些是工具说明书的可执行索引,不受文章红线约束,免检。 判据只有一条:它是「条目」还是「话」。 编号、锚点、清单项是给人查的条目,留着;一旦是讲给人听的整句叙述,就得过红线,不许有翻译腔和表演腔。一句话:目录随便编号,正文老实说话。
两种活:写好,和改写(尤其改写)
本技能干两件事,都要干好,改写这件更容易翻车。
- 写好:从头起草一段中文。相对好办——脑子里先有含义,落笔就照含义走,四条原则和红线内化进去,一次成型。
- 改写:手里已经有一段文字(常常是别人或 AI 先写好的),要把它改得像人话。这件最难,也最容易做砸。最大的坑是把改写当成"逐句翻译":AI 先吐一篇很难受的稿子,你再一句句去顺——这也是改写,效果照样烂。因为烂就烂在段落和整篇的设计上,你只在句子层面打转,怎么顺都顺不出好文章。
改写的正确起手式,永远是先看全篇:
- 读懂全文:整篇在讲什么、结构怎么搭、每段承担什么含义。
- 从含义出发定改法:这段到底想表达什么?站在全篇看,这段还需不需要存在?这句还需不需要留?
- 能大段改就别抠单句:有完整上下文、含义也吃透了,就尽量以整段甚至整篇为单位改写;越往下拆到单句,越容易只顾局部、丢了全局。
- 最后才落到句子:每段的每句该表达什么意思、该怎么改,是拆到最后一步的事,不是第一步。
为什么单句改不动那些真正的烂句:有些句子单看就是坏的——含义本身没立住,你在句子里怎么倒腾都是坏的。这时唯一的出路是放宽上下文,结合整段甚至全篇重新组织:可能这句压根不用留,可能这整段都能砍。脱离上下文孤零零改一句,你既不敢大改,也改不对,这是死路。
四条原则(全档共享,顺序即优先级)
一、保事实(最高,一个字都不能错)
改中文的过程里,原意、数字、命令、术语、谓词方向、效果类型、关系类型、硬机制描述、论文级证据,一律不许动。听着再顺,把「压力大→辞职多」改成「辞职多→压力大」就是废稿。
保真高于加中文味:精确机制(如「attention 是 Q-K 相关度加权」)不许被顶替成一个具体场景。论文结论不许简化成「有研究发现」。想加比喻,叠在机制后面,不许替换。
What ships with it
18 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.
- assets/article/01-wang-li.jpg 101 KB
- assets/article/02-modern-chinese-grammar.jpg 103 KB
- assets/article/03-lu-xun-hard-translation.png 2663 KB
- assets/article/04-gpt3-chinese-corpus.png 958 KB
- assets/article/05-rlhf-feedback-loop.png 1417 KB
- assets/article/06-europeanized-chinese-grammar.png 1094 KB
- assets/article/07-chinese-rhythm.png 1532 KB
- assets/article/08-lu-xun-speak-chinese.png 2603 KB
- README.en.md 3.0 KB
- README.md 39 KB
- references/before-after-worktext.md 6.3 KB
- references/chinese-four-principles.md 6.4 KB
- references/protected-spans.md 4.6 KB
- references/quick-scan-regex.md 6.2 KB
- references/rhythm-check.md 7.4 KB
- references/symptom-dictionary.md 17 KB
- references/word-coining-checklist.md 6.0 KB
- references/workflow.md 7.0 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.
- 12d ago First seen · 291 lines · 225 tokens per session scan A 919d95dc532e
haohao-shuohua is a skill published in the GitHub repository Job-Yang/jobbyang-ai-skills (67 stars, last pushed 9d ago), licensed MIT. It adds 225 tokens to every session and 10,685 once invoked, about $0.0011 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…