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 peterfei/forge-skill --skill occams-razor-skillgit clone --depth 1 https://github.com/peterfei/forge-skillWrote 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/peterfei/forge-skill/occams-razor-skill)<a href="https://agentmods.dev/skills/peterfei/forge-skill/occams-razor-skill"><img src="https://agentmods.dev/badge/skills/peterfei/forge-skill/occams-razor-skill.svg" alt="Measured on agentmods" 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 | $0.00100 | $0.02754 |
| Opus 5 | $0.00050 | $0.01377 |
| Sonnet 5 | $0.00020 | $0.00551 |
| Haiku 4.5 | $0.00010 | $0.00275 |
Grade A, and why
occams-razor-skill 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 — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.
奥卡姆剃刀 · 思维工具
"Entities should not be multiplied beyond necessity." — William of Ockham
激活条件与触发词
- 直接调用:「用奥卡姆剃刀」「最简解释是什么」
- 语义触发:用户面对多个复杂解释/方案无从选择,或明显在过度复杂化问题
- 组合调用:「先用奥卡姆剃刀简约化,再用系统思维看简化后是否丢失关键要素」
方法框架概览
奥卡姆剃刀的核心是在同等解释力下选择最少假设的方案,它通过假设计数、解释力对比、简单性排序三个动作来避免过度复杂化认知和决策。
核心原理(3-7个,每个须附 ≥2 个跨域证据)
原理 1: 同等解释力下选择最少假设
一句话定义:两个理论能解释同样的现象时,选假设更少的那个。
跨域证据:
- 天文学:Copernicus 日心说用更少假设解释了天体运动,替代了托勒密需要 80 多个本轮的复杂地心说(来源:Kuhn, The Copernican Revolution, 1957)
- 医学诊断:医生被训练为"如果听到蹄声,想的是马不是斑马"——先用最简假设诊断而非罕见病(来源:Groopman, How Doctors Think, 2007)
应用方式:面对多个方案时,列出每个方案的假设数。如果可以解释同样多的现象,选假设最少的。
局限:简单不等于正确。有些现象(如量子力学)确实需要非常复杂的假设来解释。
原理 2: 简单性是人类认知的启发式工具,不是真理标准
一句话定义:我们选简单的方案不是因为简单=正确,而是因为简单方案更容易被验证和修正。
跨域证据:
- 科学哲学:Popper 指出简单理论更容易被证伪,这才是简单性的真正优势(来源:Popper, Conjectures and Refutations, 1963)
- 机器学习:L1 正则化(Lasso)倾向于产生稀疏模型——不是因为稀疏模型更真实,而是因为它们更可解释且泛化能力更好(来源:Tibshirani, Regression Shrinkage via the Lasso, 1996)
应用方式:选简单方案不是因为"这就是真相",而是因为"如果错了我能更快发现并调整"。
局限:把"更可验证"误解为"更正确"是奥卡姆剃刀最常见的误用。
原理 3: 复杂假设需要更强的证据支撑
一句话定义:你想说服我相信一个复杂的解释?你需要成比例地更强的证据。
跨域证据:
- 法学:刑事审判中"排除合理怀疑"(beyond reasonable doubt)要求对复杂犯罪假设需要更多证据(来源:英美法系刑事证据标准)
- 算法选择:在机器学习模型选择中,复杂度更高的模型需要更多的训练数据来支撑(来源:Vapnik, The Nature of Statistical Learning Theory, 1995)
应用方式:如果一个方案比另一个复杂得多,问"这个复杂度带来的额外解释力值得吗?你有足够证据支撑这些额外假设吗?"
局限:有些问题(如复杂系统的行为预测)天生需要复杂模型。用奥卡姆剃刀削掉必要的复杂度就是自欺欺人。
操作协议(Agentic Protocol)
Step 1: 问题分类
适用信号:
- 面对多个可能的解释或方案不知选哪个
- 当前的解释或方案包含大量假设但看不到必要性
- 决策被过度复杂化,需要降维简化
不适用信号:
- 问题涉及复杂适应系统(如金融市场、生态系统)——简单解释通常是错的
- 处理的领域有公认的复杂理论(如量子力学)——不要用剃刀削掉物理学
- 需要精确而非简化的场景(如法律合同、医疗诊断)
Step 2: 奥卡姆剃刀式分析
-
假设枚举与计数:列出现有的所有候选方案,每个方案拆解出它依赖的假设。数一数。
- 来源原理:原理 1(最少假设)
- 操作方式:每个假设写成"要成立,需要假设______"。只数不可直接验证的假设
-
解释力对比评估:每个方案能解释用户面对的现象吗?解释到什么程度?
- 来源原理:原理 2(启发式工具)+ 原理 3(证据支撑)
- 操作方式:每个方案打分(0-10),0=完全不能解释,10=解释所有已知现象。复杂方案必须比简单方案多解释至少 2 个现象才值得考虑
-
简单性排序:按"假设数量+解释力"综合排序
- 来源原理:原理 1+2+3
- 操作方式:如果简单方案和复杂方案解释力相同→选简单方案。如果复杂方案解释力显著更高→需要你判断"这个额外解释力值得吗?"
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 · 164 lines · 100 tokens per session scan A dba303492b53
occams-razor-skill is a skill published in the GitHub repository peterfei/forge-skill (13 stars, last pushed 2mo ago), licensed MIT. It adds 100 tokens to every session and 2,754 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
opencli-usage
Use at the start of any OpenCLI session — this is the top-level map of what opencli can do, how to discover adapters, what flags and output formats are universal, and which specialized skill to load next. Point here when an agent asks "what can opencli do?" or "how do I find the right command?".
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
opencli-browser-sitemap
Use when driving a website with opencli browser and sitemap context is available, requested, or needed to avoid blind navigation. Guides agents to consume site sitemap files lazily, choose adapter/browser fallback paths, resume from state signatures, and mark stale sitemap entries without trusting them over live…
crit-story
Author a crit story only when the user explicitly invokes crit-story or directly asks you to generate a crit story. Do not infer this skill from generic review, PR, or diff-review requests.
antv-x6-editor
A skill for creating and troubleshooting interactive diagrams with AntV X6, a JavaScript engine for editors made of connected nodes and lines. It supports diagram types such as flowcharts, dependency graphs, entity-relationship diagrams, and organization charts.
gpt-vis
A chart-generation tool that recommends visual formats for data and produces either chart settings or runnable code. It uses GPT-Vis, a library for rendering data visualisations.