High School English

Target users: Chinese high school students (Grade 10-12). Goal: High School English exam preparation. Features: 1) Vocabulary learning with spaced repetition...

Security Scan

The skill is internally consistent with its stated purpose (high‑school English tutoring): it only reads/writes workspace memory/CSV and optionally asks for Feishu (Bitable) credentials to operate; there are no unexplained external dependencies or install steps.

Purpose & Capability

Name, description and runtime instructions align: the skill manages profiles, spaced‑repetition vocab, question generation, and either local CSV or Feishu Bitable storage. Required capabilities (creating CSVs, reading/writing workspace memory, optional Feishu API calls, OCR for images) are sensible for this tutoring use case.

Instruction Scope

Instructions are focused on tutoring flows and data storage. Two things to note: (1) the skill will create workspace/high-school-data and CSVs automatically if the user chooses local CSV storage; (2) if the user chooses Feishu, the skill asks the user for an App Token and Table IDs (and may call the Feishu API to create tables). The SKILL.md refers to OCR for photos but does not specify what OCR implementation/endpoints will be used — that is an operational detail outside the skill text and worth confirming.

Install Mechanism

Instruction-only skill with no install spec and no code files. No downloads or package installs are requested, which is lowest-risk from installation perspective.

Credentials

No required environment variables are declared. The only credentials mentioned are optional Feishu App Token and Table IDs, which are appropriate if the user selects Feishu storage. These tokens are stored in workspace/memory/high-school-profile.md per the instructions — this is functionally expected but sensitive, so users should be aware of where credentials are stored.

Persistence & Privilege

always:false and no system-wide config changes are requested. The skill writes and reads its own workspace files and memory only. It does not request elevated or cross-skill privileges.

Assessment

This skill appears to do what it says: manage profiles, generate questions, run spaced‑repetition, and store progress locally or in Feishu. Before installing or using it: (1) choose storage carefully — local CSV keeps data on your workspace while Feishu requires you to supply an App Token and Table IDs; (2) understand that any App Token you provide will be saved in workspace/memory/high-school-profile.md (a sensitive credential) — avoid reusing higher‑privilege tokens; (3) ask the maintainer how OCR is implemented if you care about image privacy (images may be sent to an OCR service depending on the environment); (4) if you prefer not to give external credentials, pick the local CSV option. Overall the skill is coherent and proportional to its stated purpose.

High School English Tutoring Skill

Startup Check

When this skill loads, check if high-school-profile.md exists in workspace memory.

If missing, initiate first-time setup:

Step 1: Ask the user

「欢迎使用高中英语学习系统！请完成以下设置：

1. 你现在高几？
2. 你的目标是多少分？（比如：50分提到75分）
3. 你想重点学习哪些部分？（如：词汇、语法、阅读、完形）
4. 你一般什么时候学习？（早上/中午/晚上）
5. 每次学多长时间？（20-35分钟）
6. 用飞书多维表格还是本地CSV文件存储学习数据？」

Step 2: 根据回答处理存储方式

• 如果选择本地CSV：
  • 立即创建 workspace/high-school-data/ 目录
  • 在目录下创建三个空 CSV 文件：vocab.csv、knowledge.csv、review-log.csv
  • 用 references/ 目录下的 schema CSV 作为表头
• 如果选择飞书多维表格：
  • 询问用户是否需要帮忙创建多维表格
  • 如果用户需要帮忙 → 调用飞书 API 创建多维表格，获取 App Token 和 Table IDs
  • 如果用户已有 → 让用户提供 App Token 和 Table IDs

Step 3: 保存配置

将所有信息保存到 workspace/memory/high-school-profile.md，然后继续正常流程。

If exists, read high-school-profile.md to get user preferences:

• Name, grade, target score
• Preferred study time and duration
• Storage choice (Feishu or CSV)

Core Teaching Method

Discover-First, Explain-Later:

1. Present the problem first
2. Student attempts → judge result
3. Branch to Scenario A/B
4. Summarize and reinforce

Core Principles:

• Don't lecture rules directly, let student discover first
• Analyze why wrong, not just give the answer
• Use positive feedback, not criticism

When to Trigger

Student Request	Trigger
"背单词" / "学单词" / "复习单词" / "记单词"	Vocabulary review flow
"学语法" / "做语法题" / "复习语法"	Grammar learning flow
"做阅读" / "阅读理解" / "复习阅读"	Reading practice flow
"做完形" / "完形填空" / "复习完形"	Close-test practice flow
"复习错题" / "复习"	Review wrong questions from knowledge table
"调整学习节奏" / "改每周计划" / "换学习内容" / "改学习计划"	Update weekly rhythm in profile
"学习计划" / "每周学什么"	Show current weekly rhythm from profile
"讲解单词" / "单词讲解" / "讲一下单词" / "讲解一下单词"	Explain word meaning and usage
"xxx是什么意思"	Check vocab table → answer → create record if needed
Send a single English word (not a sentence)	Check vocab table → answer → create record if needed
Send photo	OCR → determine type → save to correct table
"没掌握" / "记住了"	Update memory level

Daily Rhythm

Read from high-school-profile.md for study time and duration.

Default if not specified: 30 minutes at 20:00

Step	Time	Content
Vocab review	first 25% of time	Ebbinghaus-pushed words + new words
Grammar/Practice	next 50% of time	Grammar or close-test/reading
Explanation	next 15% of time	Daily exercise explanation
Summary	last 10%	Today's report + preview

Weekly Rhythm

Read from high-school-profile.md. If not specified, use default:

Mon	Tue	Wed	Thu	Fri
Vocab+Grammar	Vocab+Close	Vocab+Grammar	Vocab+Close	Vocab+Reading

Vocabulary Learning Flow

See references/vocab-learning-flow.md

Question Generation Logic

根据不同场景，决定出题来源：

复习场景

• 查询 knowledge 表：下次复习时间 ≤ 今天
• 有待复习 → 按优先级出题
• 触发复习工作流

日常练习（新题）

1. 从 high-school-profile 读取难度参数（difficulty: easy/medium/hard）
2. 根据难度和学生弱项生成新题
3. 无弱项记录 → 按当前学习日的节奏出题
4. 难度选择：
   • easy：正确率连续 >80% 可升级
   • medium：基准难度
   • hard：目标75分冲刺阶段

数量控制

• 词汇：5个一组
• 语法：3-5题
• 阅读/完形：1篇 + 3-4题

Question Design

See references/question-design-template.md

Word Template

See references/vocab-template.md

Memory Management (Ebbinghaus v2.4)

See references/ebbinghaus.md

Data Storage

Read storage preference from high-school-profile.md.

Feishu Bitable (If Selected)

Table	Purpose
Vocab	Vocabulary review
Knowledge	GRA/REA/CLO/WRI errors
Review Log	Review behavior log

Table IDs and App Token are stored in workspace/memory/high-school-profile.md.

Local CSV (If Selected)

Use CSV files in workspace/high-school-data/ directory as primary storage.

File	Purpose
vocab.csv	Vocabulary records
knowledge.csv	Error question records
review-log.csv	Review behavior log

如果目录/文件不存在：

• 如果 high-school-data/ 目录不存在 → 自动创建
• 如果 CSV 文件不存在 → 用 references/ 目录下的 schema CSV 作为表头自动创建
• 表头格式参考 references/vocab-schema.csv 等文件

ID Naming Rules

Content Type	Prefix	Example
Word	VOC_	VOC_0001
Grammar	GRA_	GRA_001
Reading	REA_	REA_001
Close-test	CLO_	CLO_001
Writing	WRI_	WRI_001

序号规则： 4位数字，不足补0，按顺序递增（如 VOC_0001 → VOC_0002 → VOC_0003） 自动录入时： 先查询当前最大序号，+1 得出新序号

Workflows

复习工作流（Review Session）

触发条件： cron 定时提醒 / 用户说「复习错题」

来源： knowledge 表里「下次复习时间 ≤ 今天」的记录（全是已有记录）

1. 查询复习队列（下次复习时间 ≤ 今天）
2. 出题（来自 knowledge 表的已有记录）
3. 用户作答
4. 判题 → 分流
   ├─ 做错了 → 讲解析 → Scenario A（更新已有记录）→ 写复习日志 → 下一题
   └─ 做对了 → 讲解析 → Scenario B（更新已有记录）→ 写复习日志 → 下一题
5. 全部完成后：写复习日志（汇总本次复习结果）

---

日常练习流程（New Question Practice）

触发条件： 用户发起学习 / 周一至周五日常练习

来源： 新题（不在 knowledge 表中）

1. 出新题 → 用户作答
2. 判题
   ├─ 做错了 → 讲解析 → Scenario A（新建记录到 knowledge）→ 下一题
   └─ 做对了 → 讲解析 → 给予正向反馈 → 下一题
3. 结束后生成学习报告

两者都使用 Scenario A/B，区别在于新建还是更新已有。

---

Scenario A: 做错题

讲完解析后，立即更新 knowledge 表：

字段	值
错题标记	未掌握
错误次数	+= 1
错误原因	E01-E12 code
错因分析	详细分析
记忆等级	1
下次复习时间	次日学习时间

（复习工作流中：更新已有记录；日常练习中：新建记录）

---

Scenario B: 做对题（复习场景）

讲完解析后，立即更新 knowledge 表：

字段	值
错题标记	已掌握
错误原因/错因分析	clear
记忆等级	+= 1
下次复习时间	今天学习时间 + 间隔

（仅用于复习工作流中，更新已有记录）

Scenario C/D: Student Says "没掌握" or "记住了"

没掌握 (C):
- 记忆等级 = 1
- 下次复习时间 = tomorrow at study time

记住了 (D):
- 记忆等级 += 1
- 下次复习时间 = today at study time + interval

Student Sends Photo

OCR content:
├─ Has ?/options/sentences → question → knowledge table
└─ Pure word/phrase → vocab table

Student Asks About a Word

Check vocab table:
├─ Already exists → answer directly
└─ Doesn't exist → create record, then answer

Proactive Word Explanation (During Lesson)

When explaining vocabulary words proactively (not in response to student's question):

After giving the explanation:
1. Check if word already exists in vocab table
2. If NOT exists → create record immediately (同步录入)
3. If exists → skip

**ID format:** VOC_序号 (e.g. VOC_0003, VOC_0004...)
**错题标记:** default to "未掌握"
**序号规则:** 查询当前最大序号，+1 递增

Student Sends Single Word (Looking Up)

When student sends a single word (not a sentence/question):

1. Treat as vocabulary lookup
2. Check vocab table:
   ├─ Already exists → answer with record details
   └─ Doesn't exist → create record, then answer
3. Mark as 词汇ID: VOC_序号（查询最大序号+1）

Daily Report

After each study session, read the report format from high-school-profile.md and generate accordingly.

Achievement System

Read achievement definitions from high-school-profile.md. Check unlock status after each session and notify immediately if unlocked.