SkillFoxx

Final Review skill

A skill that converts course materials into clean Markdown and prepares exam notes, practice questions and mistake analysis.

  • 113 ★ on the repository
  • Last commit
Skill

Minimal access

Why this level

  • The skill only guides the process and writes Markdown files into the project; it calls no third-party services.
All reasons and checks

lucianwhy/final-reviewAuthor:lucianwhy

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/final-review

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Minimal accessThe tool runs nothing and makes no network calls.

  • The skill only guides the process and writes Markdown files into the project; it calls no third-party services.

This is third-party code. Review the repository files before installing.

What it does

Final Review is a skill for exam preparation. It relies on the user's own materials rather than random question banks: lecture slides, past papers, homework and notes. First the agent holds an intro conversation about the exam format, where materials live, required chapters and dates and records the result in `AGENT.md`. Then the order is fixed: materials are converted to Markdown with `markitdown`, cleaned of junk and duplicates, knowledge is extracted from them and only then are questions composed. Source priority is past papers, teacher slides, homework, other exercises and, last, AI-written questions. The output is answers and walkthroughs focused on exam points: what to learn, what to write and how to earn marks. The repository also holds a separate FastAPI backend, but the skill itself can be used on its own.

Who it is for. Students before exams who want to turn slides and past papers into notes with questions and weak-spot analysis.

Good fit when

  • An exam is a week away and you need clean notes from slides and PDFs.
  • You have past papers and want new questions in the same style.
  • You need a list of topics to learn first.
  • You want questions and answers in a format close to your exam.

Not a fit when

  • There are no materials and you want preparation without sources.
  • You need a ready service with accounts and a database rather than a skill.
  • You need a legally or officially meaningful assessment of knowledge.

Example request

Prepare me for the linear algebra exam: convert slides and past papers to Markdown, clean them and write ten questions in the style of the past papers.

Limitations

The skill and its texts are written in Chinese, so for a non-Chinese user the agent must translate the instructions. Conversion depends on `markitdown`, and scans and complex formulas need manual checking. The repository declares no license. The backend in the repository is separate and needs Python 3.11, uv and Docker or PostgreSQL with pgvector. AI checking and grading can be wrong and do not replace the exam. The intro conversation is skipped when `AGENT.md` is already filled or you ask to work only in this chat.

How to disable. Remove the skill from the agent skills directory. `AGENT.md` and the converted materials stay in your project and are deleted by hand.

FAQ

Where does the work start?

With a conversation where the agent clarifies task types, where materials live, required chapters and the exam date. It records answers in `AGENT.md` and does not ask again.

What converts the materials?

`markitdown`: PPT, PDF, Word, images and notes become Markdown. Mandatory cleaning follows, and only then knowledge extraction and question writing.

Which sources matter most?

Past papers first, then teacher slides, homework, other exercises and quick courses. AI-written questions come last.

Do I need the repository backend?

No. The FastAPI backend with PostgreSQL and React is for a separate service, while the skill works on its own through `SKILL.md`.

Can I get HTML instead of text?

Yes. In the conversation you can choose the answer format, including plain text and HTML, and also a mock exam.

Discussion

No questions or reviews yet.

Skills

An exam prep skill built on your own lectures and problems: the agent goes through the course chapter by chapter with citations

SkillExtended access381

A tutoring skill for STEM and AI: it finds the knowledge gap first, then teaches one step and checks understanding

SkillMinimal access267

A skill for exam cramming in one or two weeks: it builds a topic tree, teaches, tests and patches gaps.

SkillMinimal access180

A tutor for any subject: diagnosis, a study plan, project tasks and mastery checks, with the results saved to Markdown files

SkillExtended access205

A skill for exam and contest prep: topic checklist, day-by-day plan, daily lessons and a mistake notebook built from real answers.

SkillExtended access326

README in short

The repository grew from a skill into a study backend with material upload, sourced Q&A, quizzes and weak-spot feedback. The original `SKILL.md` stays as the domain specification.

README badge

Original SKILL.md

---
name: final-review
description: 用于期末复习、考前突击、题目生成、知识点辅导、复习资料清洗等场景。基于用户上传课程资料开展工作，优先参考历年真题、老师PPT、平时作业和 Markdown 资料；先用 markitdown 转 Markdown，再清洗材料，最后生成题目、答案和偏考试得分导向的解析。
---

# 期末复习 Skill

把用户资料整理成偏考试导向的笔记、题目和解析。

## 引导对话

Skill 被调用后，**第一步不是开始转换材料或出题，而是先和用户对话一轮**，把考试信息和资料源信息收集清楚，写入 `AGENT.md`，以后不复述。

### 触发时机

- 用户触发 final-review skill
- 当前 `AGENT.md` 中考试信息为空或标注"待补充"
- 用户已明确表示进入复习场景

### 不触发引导的情况

- `AGENT.md` 中已有完整信息（不标注"待补充"）
- 用户说"这次不改 AGENT.md"或"只在这个对话里做"
- 用户已经在前一轮对话中说过

### 引导问题清单

按顺序问，用户说过的不重复问：

1. **考试题型**：这门课考什么题型？有题型样例或考试范围吗？
2. **资料位置**：复习资料放在哪些位置？每份是什么来源？（授课 PPT、课后作业、速成课、历年真题…）
3. **重点章节**：老师强调的必考/重点章节有哪些？有不考的范围吗？
4. **考试日期**：什么时候考？（可选，方便排序和提醒）
5. **用户偏好**：输出格式偏好（文本 / HTML）、要不要模拟考试等。（可选）
6. **md 格式资料路径**：仓库里已经有的 `.md` 格式复习资料，分别在哪些路径？

问完一轮后，**把收集到的信息整理展示出来，让用户确认无误后再写入 `AGENT.md`**。写入后在回复里简要说明记了什么。

## 核心规则

- 先基于用户提供资料工作，不依赖随机外部题库。
- 默认用 `markitdown` 把资料转成 Markdown。
- 必须先清洗 Markdown，再提知识点、再出题。
- 目标是考试表现：背什么、写什么、怎么拿分。

Report a problem