Feynman logo

Feynman

Community
fainshare
feynman

Audit(审查·太虚四转·破妄): six self-deception principles via blind sub-agent. Use when user says "审查", "费曼", "feynman", "提交前把关", "反自欺", "/feynman".

Overview

Publisherfainshare
RepositoryLiam-Skills
Skill namefeynman
Stars
113
Forks
19
Bundled files
5
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by fainshare on GitHub. Read the source before you install it.

Installation

Install the Feynman AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/fainshare/Liam-Skills/tree/main/skills/feynman
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/fainshare/Liam-Skills.git /tmp/Liam-Skills
mkdir -p .claude/skills
cp -r /tmp/Liam-Skills/skills/feynman .claude/skills/feynman
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Feynman in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Feynman on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Feynman is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Feynman 费曼审查

"The first principle is that you must not fool yourself — and you are the easiest person to fool." — Richard Feynman

概述

代码写完、PRD 交稿,方案落地。但里面有没有在骗自己?AI 倾向于自圆其说,会说"这样对"而不是"可能有问题"。这个 skill 用六条原则检查产出物:假设没验证就当事实用、猜测被包装成结论,为不存在的需求写了代码、顺手改了不该改的地方,用想象代替实验、制造真相副本而不是引用原始来源。审查员是独立子 Agent,只看变更产物,看不到对话历史,没法被你的理由说服。

太虚五转第四转:破妄(审查)。提交前跑一遍六原则,把自欺的部分找出来。

理论框架

六条原则构成一条认知链:停下来(P1)→ 锚定推理(P2)→ 设计要简(P3)→ 执行要准(P4)→ 验收要闭环(P5)→ 数据要溯源(P6),分别防止假设伪装、伪装确定性,过度设计、附带伤害,自证清白、真相漂移。

原则权威定义:CLAUDE.md(SSOT,运行时动态读取,不内联)。

执行模式:subAgent 独立审查(默认)

同一个 Agent 既写代码又审查代码,上下文里全是"为什么这样做"的理由——你在用已有的理由为自己辩护,不是在审查。subAgent 隔离的是上下文,不是能力。 审查者只拿到 diff + 原则 + checklist,看不到对话历史、用户请求,你的推理过程。

Phase 1(收集变更)在主 Agent 完成,Phase 2 + Phase 3,审查 + 报告委派给 subAgent。主 Agent 向 subAgent 传递数据包(diff_content / changed_files / principles / checklist / report_template / problem_type),禁止传递对话历史、推理过程,用户评价、"为什么这样改"的解释。详见 references/subagent-protocol.md。

模式选择、降级条件,problem_type 适配、--scope 模式说明,与 liam-retro 分工详见 references/decision-tables.md。

输入

默认审查当前 git 工作区的未提交变更(git diff + git diff --cached)。也接受:指定文件路径、指定 commit 范围,指定 MR/PR、存量审查(--scope module)、依赖链审查(--scope deps)、--problem-type,可选,pipeline 传入。

审查流程

Phase 1 收集变更(主Agent)──→ 委派 subAgent ──→ Phase 2+3 审查+报告(subAgent)
     │                            │                        │
  输入: diff                  隔离上下文             输入: diff+原则+checklist
  输出: 文件清单+分类         只传数据包             输出: 审查报告 markdown
     │                                                    │
  CHECKPOINT 确认范围                              参照路由矩阵按文件类型聚焦

Phase 1: 收集变更

收集 diff(默认)、module,指定目录代码文件、或 deps,diff 函数的 1 层调用链。超过 50 个文件(module)或 20 个依赖文件(deps)时要求用户缩小范围。

CHECKPOINT:展示变更文件清单({N}个文件,{+lines}+/{-lines}-),确认审查范围后继续。

文件类型×原则路由矩阵、边界条件,module/deps 审查聚焦策略详见 references/decision-tables.md。

Phase 2: 逐原则审查

对每条原则,按路由矩阵决定审查深度:● 逐条过 checklist,○ 快速扫描。判定 ✅ 通过 / ⚠️ 警告 / ❌ 失败。必须引用具体证据 —file_path:line_number + 代码片段或 diff 行。

判定分级:❌ 判定区分"技术否决"(代码有明确错误)和"判断否决",推翻设计决策。判断否决必须附带可执行的反例验证步骤,30 秒内可跑、可证伪,声明预期。附不出则降级为 ⚠️ 并标注 [待验证]。

详细检查项见 references/checklist.md。

不确定点深挖(Bayes 联动)流程详见 references/decision-tables.md 深挖章节。subAgent 深挖 prompt 模板详见 references/subagent-protocol.md。

Phase 3: 输出报告

报告含 YAML frontmatter + 审查工作量量化 + 逐项判定表(6原则 × ✅/⚠️/❌)+ 合规率 + 问题清单 + 总结 + 下一步引导。判定阈值:6/6 通过,有⚠️无❌通过带建议,有❌不通过。详见 references/report-format.md。

不做的事

  • 不做安全审计(用 liam-cso skill)
  • 不做功能正确性测试(用单元测试 / e2e 测试)
  • 不做代码风格检查(用 linter)
  • 不修复问题——只报告。用户确认后再修
  • 默认(diff 模式)不审查存量代码,除非用户指定 --scope module/deps
  • 不审查会话行为模式(用 liam-retro Step 6)

沉淀协议:写入 BACKLOG

审查完成后有 ⚠️ 或 ❌ 级发现时,用 CLI flag 模式调用 sink-to-backlog.py(❌→ERROR, ⚠️→WARNING, ✅→跳过)。详见 references/report-format.md。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Feynman AI skill do?

Audit(审查·太虚四转·破妄): six self-deception principles via blind sub-agent. Use when user says "审查", "费曼", "feynman", "提交前把关", "反自欺", "/feynman".

Why use Feynman on TypingMind?

Because you install it once and use it with any model. Feynman is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Feynman in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/fainshare/Liam-Skills/tree/main/skills/feynman. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Feynman?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Feynman?

As many as you like. As long as a model supports skills, you can use Feynman with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Feynman AI skill free?

It is published on GitHub by fainshare. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇