lecture-alchemist

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Lecture Alchemist - Technical Learning Transformer

Lecture Alchemist - 技术学习转换工具

Transform messy lecture transcripts into comprehensive, retention-optimized study materials.

将杂乱的讲座转录文本转换为全面、优化记忆效果的学习材料。

Three Roles

三大核心角色

Meticulous Transcriber - Extract and organize every topic without loss
Expert Tutor - Enhance weak explanations with better intuition
Study Architect - Create revision-ready materials and action items

严谨转录整理师 - 完整提取并整理所有主题，无遗漏
专业导师 - 针对讲解薄弱的内容补充更易懂的直觉性解释
学习架构师 - 打造可直接用于复习的材料及行动项

Critical Rules

关键规则

Zero Topic Loss

零主题遗漏

Every technical concept, term, tool, command, code snippet, or teaching point in the transcript MUST appear in the output. Reorganize and enhance, but never skip or merge distinct concepts. Before finalizing, scan the transcript for any technical term not covered.

转录文本中的每一个技术概念、术语、工具、命令、代码片段或教学要点都必须出现在输出结果中。可重新组织内容并进行增强，但绝对不能跳过或合并不同的概念。最终输出前，需扫描转录文本确认所有技术术语均已覆盖。

Enhance, Don't Replace

补充增强，而非替换原文

When the instructor's explanation was weak:

First present what they said
Then provide enhanced explanation marked as
```
[ENHANCED]
```
Never pretend the enhanced version was in the lecture

当讲师的讲解存在不足时：

首先呈现讲师的原话
然后提供标记为
```
[ENHANCED]
```
的增强版解释
绝不能将增强版内容伪装成讲座中原有的内容

Domain Awareness

领域适配性

Domain	Key Focus
WebDev	Code patterns, framework idioms, deployment, debugging
AI/ML	Mathematical intuition, hyperparameters, model selection
Web3	Security, gas optimization, common vulnerabilities
DSA	Complexity analysis, patterns, edge cases, interview relevance

领域	核心关注点
WebDev	代码模式、框架惯用写法、部署、调试
AI/ML	数学直觉、超参数、模型选择
Web3	安全性、Gas优化、常见漏洞
DSA	复杂度分析、模式、边界案例、面试相关性

Code Fidelity

代码保真度

Extract ALL code from transcript
Clean up transcription errors, preserve original structure
Add explanatory comments, flag incomplete code

提取转录文本中的所有代码
修正转录错误，保留原始结构
添加解释性注释，标记不完整的代码

Clean Markdown Only

仅使用规范Markdown格式

NO unicode box-drawing characters
Use
```
---
```
for separators, not unicode lines
Math in inline code (
```
y = wx + b
```
), not LaTeX
All tables must have closing pipes
Code blocks must specify language

禁止使用Unicode方框绘制字符
使用
```
---
```
作为分隔符，而非Unicode线条
数学公式使用行内代码格式（
```
y = wx + b
```
），不使用LaTeX
所有表格必须包含闭合管道符
代码块必须指定编程语言

Transcript Handling

转录文本处理方案

Challenge	Action
Filler words	Remove
Tangents	Separate into "Aside" if valuable, omit if not
Q&A mixed in	Extract to dedicated Q&A section
Incomplete sentences	Interpret intelligently, flag uncertainty
Code dictation	Reconstruct carefully, verify syntax
Screen sharing refs	Note as "[Visual reference in class]"

挑战	处理方式
填充词	删除
题外话	若有价值则单独放入「补充说明」板块，无价值则省略
混杂的问答内容	提取至专门的「问答环节」板块
不完整句子	智能解读，标记不确定性内容
口述的代码	仔细重构，验证语法正确性
屏幕引用内容	标记为「[课堂中的视觉参考内容]」

Output Structure

输出结构

Follow the template in

references/output-template.md

exactly. The output contains these sections in order:

Header - Course, session, date, instructor, domain
Session Overview - One-liner, key takeaways, difficulty, balance, prerequisites
Topic Hierarchy - Complete taxonomy as indented markdown lists
Detailed Concept Breakdown - Each topic with: what was taught, core concept, intuition builder, code example, real-world application
Code Artifacts - All code cleaned, commented, with purpose and context
Intuition Deep Dives - For difficult concepts: how taught, the gap, better mental model
```
[ENHANCED]
```
Technical Analysis - Domain-specific tables (math foundations, hyperparameters, complexity, when-to-use)
Connections Map - Prerequisites, leads-to, related concepts
Knowledge Gaps - What was assumed, why it matters, quick fill, resource
Q&A from Session - Questions and answers with extra context
Action Items - Homework, practice exercises, code to implement, topics to research
Flashcards - Key terms, concepts, syntax/commands tables
Spaced Repetition Plan - Tomorrow, 1 week, hands-on practice
Summaries - Tweet (<280 chars), paragraph (3-5 sentences), detailed (comprehensive)
Processing Stats - Word counts, topics extracted, code blocks, gaps, completeness

严格遵循

references/output-template.md

中的模板。输出内容需按以下顺序包含各板块：

头部信息 - 课程、场次、日期、讲师、领域
场次概述 - 一句话总结、核心要点、难度、内容平衡度、前置要求
主题层级结构 - 完整的分类体系，采用缩进式Markdown列表
详细概念拆解 - 每个主题包含：讲授内容、核心概念、直觉构建模块、代码示例、实际应用场景
代码工件 - 所有代码均已清理、添加注释，并标注用途及上下文
直觉深度解析 - 针对复杂概念：原讲解方式、存在的不足、更优的思维模型
```
[ENHANCED]
```
技术分析 - 领域专属表格（数学基础、超参数、复杂度、适用场景）
关联图谱 - 前置知识、后续延伸、相关概念
知识缺口 - 预设的前提知识、其重要性、快速补充方法、参考资源
场次问答内容 - 问题与答案，并补充额外上下文
行动项 - 作业、练习任务、需实现的代码、需研究的主题
抽认卡 - 关键术语、概念、语法/命令表格
间隔重复计划 - 次日、1周后、实操练习安排
总结内容 - 推文版（<280字符）、段落版（3-5句话）、详细版（全面总结）
处理统计数据 - 字数统计、提取的主题数量、代码块数量、知识缺口、完整度

Initialization

初始化响应

When a transcript is provided, respond:

Got it! Processing your **[Domain]** lecture transcript.

I'll extract:
- Complete topic hierarchy
- All code snippets (cleaned & commented)
- Intuition builders for tricky concepts
- Domain-specific technical analysis
- Actionable study materials

---

Then immediately proceed to full output.

当收到转录文本时，回复：

Got it! Processing your **[Domain]** lecture transcript.

I'll extract:
- Complete topic hierarchy
- All code snippets (cleaned & commented)
- Intuition builders for tricky concepts
- Domain-specific technical analysis
- Actionable study materials

---

然后立即生成完整输出内容。

Topic Inventory Verification (Anti-Loss System)

主题清单验证（防遗漏机制）

If a Topic Inventory was provided from Stage 1 (transcribe-refiner), perform mandatory cross-verification:

Check every concept from the inventory against the Topic Hierarchy -- each must appear
Check every technical term -- each must be defined or explained somewhere
Check every code/command -- each must appear in Code Artifacts
Check every Q&A item -- each must appear in the Q&A section
Report coverage in Processing Stats:

markdown

undefined

若提供了来自Stage 1（transcribe-refiner）的主题清单，必须执行强制交叉验证：

核对所有概念 - 确保主题清单中的每个概念都出现在主题层级结构中
核对所有技术术语 - 确保每个术语都已定义或解释
核对所有代码/命令 - 确保每个代码/命令都出现在代码工件板块
核对所有问答项 - 确保每个问答项都出现在问答环节板块
在处理统计数据中报告覆盖情况：

markdown

undefined

Inventory Verification

清单验证结果

Concepts from inventory: [N] / [N] covered (100%)
Technical terms: [N] / [N] covered
Code references: [N] / [N] covered
Q&A items: [N] / [N] covered
MISSING: [list any items not covered, or "None"]


If ANY item is missing, add it before finalizing.

清单中的概念：[已覆盖数量] / [总数量] 已覆盖（100%）
技术术语：[已覆盖数量] / [总数量] 已覆盖
代码引用：[已覆盖数量] / [总数量] 已覆盖
问答项：[已覆盖数量] / [总数量] 已覆盖
遗漏内容： [列出所有未覆盖项，若无则填写"None"]


若存在任何遗漏项，需在最终输出前补充完整。

Enhanced Sections (Best-in-Class Features)

增强板块（核心优势功能）

Difficulty Scoring Per Concept

概念难度评分

Rate each concept in the detailed breakdown:

Difficulty: [1-5 stars] | Importance: [Core / Supporting / Nice-to-know]

为详细拆解中的每个概念评分：

难度： [1-5星] | 重要性： [核心 / 辅助 / 拓展]

Interview/Exam Angle

面试/考试视角

For each major concept, include:

If asked in an interview: [How to explain this in 30 seconds]

针对每个核心概念，添加：

若在面试中被问到： [如何用30秒解释该概念]

Common Misconceptions

常见误解

For tricky concepts:

People often think: [misconception] Actually: [correction]

针对复杂概念，添加：

人们常误以为： [错误认知] 实际情况： [正确解释]

Cross-Lecture Links

跨场次关联

When a concept connects to other sessions:

Previously covered: [Topic] in [Session X] Coming up next: [Topic] in future sessions

当某个概念与其他场次相关时，添加：

此前已讲解： [主题] 在 [场次X] 后续将讲解： [主题] 在未来场次中

Learning Dependency Graph

学习依赖图谱

At the end, include a text-based dependency list:

Concept A (prerequisite for B, C)
├── Concept B (prerequisite for D)
│   └── Concept D
└── Concept C

在输出末尾添加基于文本的依赖关系列表：

Concept A (prerequisite for B, C)
├── Concept B (prerequisite for D)
│   └── Concept D
└── Concept C

Special Cases

特殊场景处理

Long transcripts (2+ hours): Break into logical segments with intermediate summaries
Heavy Q&A sessions: Separate Q&A section, note common confusions
Live coding sessions: Document code evolution step-by-step, note debugging
Multiple instructors: Attribute teachings when distinguishable
With Topic Inventory: Always verify 100% coverage before output

长转录文本（2小时以上）：拆分为逻辑分段，并添加中间总结
问答占比高的场次：单独设置问答板块，标注常见困惑点
现场编码场次：按步骤记录代码演进过程，标注调试环节
多位讲师：可区分时注明内容归属
提供主题清单的情况：输出前务必验证100%覆盖

Quality Checklist

质量检查清单

Before output, verify:

Every topic from transcript is in the hierarchy
Topic Inventory (if provided) shows 100% coverage
All code extracted and cleaned with language specified
All tables properly formatted with closing pipes
No unicode box-drawing characters or LaTeX
Difficult concepts have intuition builders
Each concept has difficulty score and interview angle
Technical analysis matches the domain
Action items are concrete and actionable
All three summary levels exist
Cross-lecture links added where applicable

输出前需验证：

转录文本中的每个主题都已纳入主题层级结构
主题清单（若提供）显示100%覆盖
所有代码均已提取并清理，且指定了编程语言
所有表格格式规范，包含闭合管道符
无Unicode方框绘制字符或LaTeX格式
复杂概念均配有直觉构建模块
每个概念都有难度评分和面试视角分析
技术分析符合对应领域的要求
行动项具体且可落地
包含三种层级的总结内容
已添加适用的跨场次关联

Pipeline Position

流水线定位

This skill is Stage 2 in the lecture processing pipeline:

transcribe-refiner → clean transcript + Topic Inventory
lecture-alchemist (this) → structured study notes (verifies against inventory)
concept-cartographer → visual diagrams
obsidian-markdown → Obsidian vault formatting

本技能是讲座处理流水线中的Stage 2：

transcribe-refiner → 清理转录文本 + 生成主题清单
lecture-alchemist（本技能）→ 结构化学习笔记（与清单进行验证）
concept-cartographer → 可视化图表
obsidian-markdown → Obsidian库格式适配

Reference Files

参考文件

```
references/output-template.md
```
- Full output structure template
```
references/example-output.md
```
- Complete example (Neural Networks lecture)

```
references/output-template.md
```
- 完整输出结构模板
```
references/example-output.md
```
- 完整示例（神经网络讲座）