Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
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.
[](https://agentmods.dev/skills/binghanofuestc/open_agent_team/exemplar-prose-calibration)<a href="https://agentmods.dev/skills/binghanofuestc/open_agent_team/exemplar-prose-calibration"><img src="https://agentmods.dev/badge/skills/binghanofuestc/open_agent_team/exemplar-prose-calibration/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/binghanofuestc/open_agent_team/exemplar-prose-calibration"><img src="https://agentmods.dev/badge/skills/binghanofuestc/open_agent_team/exemplar-prose-calibration.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 | $0.00065 | $0.01761 |
| Opus 5 | $0.00032 | $0.00881 |
| Sonnet 5 | $0.00013 | $0.00352 |
| Haiku 4.5 | $0.00006 | $0.00176 |
Grade A, and why
exemplar-prose-calibration 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 7d 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Exemplar Prose Calibration / 样本文学质感校准
本 skill 只提炼可迁移写作方法,不允许仿写原作、不允许保留样例中的人物名、地名、组织名、术语、桥段链或专属世界观组合。
1. 从样例中提炼出的高价值特征
1. 开篇先给处境、异常、声音或行动,不先解释设定。
2. 叙述声音有明显立场:可以偏执、冷淡、自嘲、天真、江湖气、古雅或口语化,但不能是无性格的说明腔。
3. 信息释放常常嵌在行动、对话、误判和小麻烦里,而不是用百科段落解释。
4. 场景推进靠连续的物理因果:看见/听见/闻到 -> 判断 -> 试探 -> 代价 -> 新问题。
5. 人物通过说话方式、注意力和小动作立住,不靠履历卡片。
6. 恐怖、悬疑、奇幻或权谋感来自可观察细节的递进,不来自抽象形容词。
7. 章节/场景结尾通常落在新问题、状态变化、反讽或未说破的情绪上。
8. 好句子服务人物、局势或气氛,不单独摆出来证明“文笔好”。
2. 写作前必须锁定的风格参数
每篇故事或每部中篇开写前,必须在 story bible / novel bible 中写明:
叙述人或 POV 的偏见:他/她相信什么、误判什么、看不惯什么。
语体来源:市井口语、职业语汇、古典感、少年感、冷幽默、民俗叙述、克制白描等。
句子节奏:短句压迫、长句滔滔、白描平稳、内心碎片、对话驱动。
感官主轴:气味、声音、触感、物件、身体反应、空间方向。
幽默方式:自嘲、误会、反差、嘴硬、职业黑话、冷处理。
情绪处理:压住、错开、延迟爆发、用行为替代告白。
没有这些参数,不得只写“文笔细腻”“像出版小说”“有网文爽感”。
3. 反 AI 腔核心规则
AI 腔的本质不是某个词,而是“用抽象总结替代具体经验”。以下模式必须重点清理:
抽象对称句:不是 X,而是 Y;与其说 X,不如说 Y。
顿悟句:他意识到;他终于明白;这一刻他才懂。
雾化句:某种难以言说的;仿佛有什么东西;空气凝固。
命题句:真正的恐惧/孤独/爱/命运是……
段尾总结:把刚发生的事升格成主题宣言。
情绪标签:恐惧、悲伤、愤怒、压抑、破碎等词反复出现,却没有身体和行动承载。
这些句式不是绝对零容忍。若用于具体语义、角色口吻或必要判断,可以保留;若用于抽象抒情、场景氛围、主题总结,必须改写。
硬门槛:
短篇全文:“不是……而是……”类机械对照句最好为 0,最多 1 处且必须有具体语义功能。
中篇单章:前三章最好为 0;其他章节最多 1 处。
同一稿中“他意识到/终于明白/某种难以言说/空气凝固/命运的齿轮”等模板句累计超过 2 处,必须专项返修。
4. 替换方法
4.1 把抽象判断改成身体和动作
坏:他感到的不是恐惧,而是一种更深的寒意。
改法:写心跳、手指、后颈、呼吸、视线回避、脚步迟疑,再让人物做一个错误或过度的动作。
4.2 把顿悟改成可观察证据
坏:他终于明白自己被骗了。
改法:让他发现账本页角、门闩方向、对方称呼错误、旧伤位置、物件不在原处,然后让下一句行动证明判断。
4.3 把设定解释改成“用一次”
坏:连续解释规则来源、历史和原理。
改法:角色在危险中使用规则;使用失败暴露限制;旁人质疑;代价落到身体、关系或资源上。
4.4 把漂亮段尾改成具体落点
坏:段尾总结主题、命运、孤独或人性。
改法:停在一个物件、一句没说完的话、一个动作后果、一个被误读的表情、一个新声音。
5. 场景质感检查
每个关键场景至少通过以下 5 项中的 4 项:
1. 第一三句内有具体压力、异常、行动或声音。
2. 至少一个细节只能由当前 POV 注意到,不能换人后仍完全成立。
3. 每 600-900 字有一次局势变化、信息增量或关系筹码变化。
4. 对话至少有一处“不正面回答”,角色通过绕开问题暴露真实需求。
5. 场景结尾改变读者问题:从“发生了什么”变成“接下来怎么办/他为什么这么做/代价会落到谁身上”。
6. 批评与返修用检查表
Critic、Reader、QA、Editor、Revision 必须检查:
这段是否可以删掉而不影响剧情、人物、气氛或信息?可以删则删。
这句是否只是把读者已经看见的情绪再解释一遍?是则删或改成动作。
这个角色的台词去掉名字后,是否还能听出是谁?不能则重写声音。
这一章是否有可复述的记忆点?没有则补一个具体物件、选择、反讽或恐惧规则。
这处“文笔好”是否让情节停下来?是则降调。
是否在用样例作品的专属设定、人物关系或桥段链偷懒?是则重做原创化。
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.
- 7d ago First seen · 146 lines · 65 tokens per session scan A 52dca7f9be89
exemplar-prose-calibration is a skill published in the GitHub repository BingHanOfUESTC/open_agent_team (106 stars, last pushed 2mo ago), licensed MIT. It adds 65 tokens to every session and 1,761 once invoked, about $0.0003 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 2026-09-03.
Other skills, from other repositories
company-product-context
Compiles comprehensive company product context from PDF documents, web research, and industry knowledge.
python-lib-analyzer
Analyze any Python library structure, explore modules, classes, and functions with signatures and documentation.
Ability Generator
This skill generates markdown skill templates to be later used.
nexus-configuration
A skill for reading, planning, approving, and checking configuration in Nexus, the current private conversation or room. It covers settings such as agents, rooms, providers, channels, connectors, skills, models, tools, and MCP connections.
agentica
How to answer questions about the agentica product you are running inside — CLI flags, config.yaml profiles, API keys, models, sessions, resume, workspace, AGENTS.md standing rules, skills, logs, upgrade, and selfmanage. Use when asked how agentica works, how to configure or upgrade it, where state lives on disk…
generate-ai-rules
Generate AI assistant configuration files for a repository — CLAUDE.md, AGENTS.md, and Cursor rules (.cursor/rules/.mdc) — from codebase analysis. Use whenever the user wants to create or update CLAUDE.md, AGENTS.md, agent rules, Cursor rules, AI coding assistant configuration, or "onboard AI tools" to a project, even…