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 R2h1/my-book-skills --skill v07git clone --depth 1 https://github.com/R2h1/my-book-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/r2h1/my-book-skills/v07)<a href="https://agentmods.dev/skills/r2h1/my-book-skills/v07"><img src="https://agentmods.dev/badge/skills/r2h1/my-book-skills/v07/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/r2h1/my-book-skills/v07"><img src="https://agentmods.dev/badge/skills/r2h1/my-book-skills/v07.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 | $0.00105 | $0.01874 |
| Opus 5 | $0.00053 | $0.00937 |
| Sonnet 5 | $0.00021 | $0.00375 |
| Haiku 4.5 | $0.00011 | $0.00187 |
Grade A, and why
概率博弈决策框架 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 11d 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 — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
概率博弈决策框架
R — 原文
"玩好扑克的方法很简单,只要排除情绪的干扰,根据概率押注即可。人生亦是如此。如果有赢的机会,就要战胜规避损失倾向,大胆下注。"
— 金承焕, 第六章
I — 方法论骨架
概率博弈是将扑克思维扩展到整个人生决策的框架。核心思想:
- 人生是超长期重复博弈——单次输赢不重要,重要的是每次决策的期望值
- 只押注概率,不押注情绪——胜率 > 50% 就下注,输了不后悔('押得很好,只是概率原因')
- 克服Kluge干扰——规避损失倾向(怕亏)让人不敢下注,赌徒谬误(觉得'该轮到我赢了')让人过度下注,都需识别并排除
- 过程导向而非结果导向——好的决策可能带来坏结果(小概率事件),坏的决策可能带来好结果(运气)。关键看决策过程,而非单次结果
核心思维转换: 从'这个决定会不会输?'转向'这个决定的期望值是多少?'
A1 — 书中的应用
案例 1: 作者打扑克
- 问题: 27岁和哥哥朋友打扑克,第一天输光
- 方法论的使用: 去图书馆读3本扑克书,学会基于概率分析下注。胜率55%就下注,输了也不动摇因为'押得很好'
- 结论: 不依赖情绪,纯粹基于概率
- 结果: 初出茅庐就战胜了打扑克5年的老手
案例 2: 作者做YouTube决策
- 问题: 面对'已经是红海''会被骂'等恐惧
- 方法论的使用: 用概率思维评估——成功的概率虽不确定但>0,不做的概率是100%维持现状。胜率>0就是值得下注
- 结论: 概率思维让他克服了规避损失倾向
- 结果: 成为顶级自我开发YouTuber
案例 3: 股票投资中的情绪化决策
- 问题: 大多数人在股市中做情绪化决策(恐慌抛售、追高)
- 方法论的使用: 反面案例——被Kluge操控,用赌徒谬误('这次该轮到我了')和损失厌恶(想翻本)不断加注
- 结论: 背离概率原则下注,长期必然亏损
A2 — 触发场景
用户会在什么情境下需要这个 skill?
- 用户面临职业/投资/创业等重大决策,被恐惧或贪婪绑架
- 用户连续失败后情绪失控,想'翻本'或'放弃'
- 用户纠结于'如果失败了怎么办'
语言信号
- '如果失败了怎么办?'
- '我已经坚持了这么久,不能放弃'
- '这次该轮到我了吧?'
- '我不能再输了'
- '太冒险了'
与相邻 skill 的区分
- 与
五子棋理论的区别: 概率博弈用于'给定选项后如何下注'(执行层面),五子棋用于'选择哪个方向'(战略层面) - 与
Kluge识别与对抗法的关系: Kluge识别是概率博弈的前置步骤——必须先识别出情绪干扰,才能做理性概率判断
E — 可执行步骤
当 skill 被激活后,agent 应按以下步骤执行:
-
识别情绪干扰
- 引导用户识别当前决策中的情绪:害怕损失?想翻本?要面子?愿望思维?
- 帮助用户将这些情绪标记为'Kluge正在作怪'
- 完成标准: 用户能清晰说'我的XX情绪正在影响这个决策'
-
计算概率和期望值
- 帮用户列出所有选项及其可能结果
- 对每个结果粗略估算概率和收益/损失
- 计算期望值 = 胜率 × 收益 - 败率 × 损失
- 完成标准: 每个选项都有粗略的期望值评估
-
做出决策并放下结果
- 选择期望值最高的选项
- 明确告知:'如果输了,不是你的决策有问题,是小概率事件'
- 帮用户预设'如果输了,我该怎么应对?'(避免事后情绪失控)
- 完成标准: 用户做出了选择,且能接受'输了也是合理的结果'
B — 边界
不要在以下情况使用此 skill
- 单次生死攸关的决策(概率博弈需要多次重复才有意义)
- 信息严重不足无法估算概率时(需要先收集信息)
- 用户面临的是纯情感问题(分手、丧亲)而非决策问题
作者在书中警告的失败模式
- ce16: 概率博弈中的情绪化决策——被愤怒、愿望、自尊心绑架
- ce10: 规避损失倾向——害怕亏损而不敢下注
- ce12: 情感试探——被情绪绑架做出非理性决策
作者的盲点 / 时代局限
- 作者假设用户有能力粗略估算概率——实际上大多数人的概率直觉很差
- 现实生活中很多决策无法量化成败概率(信息不完备),套用概率博弈可能导致过度简化
- 作者没有讨论'下注频率'——即使是正期望值的下注,过于频繁也会因为波动导致破产(凯利公式缺失)
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.
- 11d ago First seen · 144 lines · 105 tokens per session scan A ddeaf5f04aa5
概率博弈决策框架 is a skill published in the GitHub repository R2h1/my-book-skills (2 stars, last pushed 26d ago), licensed MIT. It adds 105 tokens to every session and 1,874 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 2026-08-31.
Other skills, from other repositories
en-zh-translation-polish
An English-to-Chinese translation and editing process focused on natural Chinese rather than word-for-word wording.
shortfilm-prompt
Generate cinematic AI shortfilm prompts (works with Seedance 2.0, Xiaoyunque, Sora, Kling, Jimeng, Veo) using the 5-stage structure from Mx-Shell's Zombie Scavenger. Trigger when the user wants transformation sequences, multi-shot narrative shorts, weapon-charge/combat segments, emotional family/pet/farewell…
multi-agent-cn
A Chinese-language system design that makes one main agent assign tasks to five persistent subagents instead of doing the work itself.
gpt-image-2-prompt-engine
A structured prompt-writing tool for GPT-Image-2, using reusable JSON templates and a four-step workflow. It is designed for controlled commercial images such as product photos, posters, infographics, brand visuals, UI screenshots, and character designs.
simp
A relationship-advice skill that helps interpret signals, plan respectful approaches, and write sincere messages for someone you like.
chinese-write-checker
A Chinese-language review process for checking an article from an editor's, reader's, writer's, and publishing platform's viewpoints.