360 Secure SkillHub
360 安全技能中心

Daily Tutor

该 Skill 读取用户自备的学习数据文件,按配置或自动识别的字段挑选若干条尚未学习的条目输出,并记录已学条目以避免重复,适用于词汇、公式、历史事件等任意学科内容;若环境中存在 quizbuild 工具,还可据此生成一份练习题。输出内容本身为条目列表,需由调用方或模型进一步讲解。

buiphuc
buiphuc
热度 41.20
最近更新 2026-05-11 17:41:13
分类:知识依赖:代码执行 / 读写文件 / MCP 服务应用场景:教学辅导 / 测验 / 语言学习

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add daily-tutor
Metadata
name
daily-tutor
version
1.0.4
description
Get new study items for any subject. Use when: user asks for a daily lesson or new learning content. Supports any data format (language vocab, math formulas, history events, etc.). Automatically tracks learned items so they never repeat.
metadata
{"openclaw":{"emoji":"📚","requires":{"bins":["python3"]}}}

Daily Tutor Skill

📖 For full user instructions, data setup guides, and Quizbuild MCP configurations, please read the Official README on GitHub.

This skill retrieves new learning items from the user's study list (data/data.json) that they have not learned yet. It works with any subject — the data structure is automatically detected.

When to Use

✅ USE this skill when:

  • The user asks for their daily lesson.
  • The user wants to learn new vocabulary, formulas, concepts, or any study content.
  • The daily cron job asks for new items.

Data Format

Place your study data in data/data.json as a JSON array. Each item can have any fields you want — the script will auto-detect and display them all. There is no fixed schema; you define the fields that make sense for your subject.

For example formats, see references/EXAMPLES.md.

Optional Configuration

Edit data/config.json to customize behavior (all fields optional):

{
  "primary_key": "word",
  "num_items": 10,
  "subject_name": "Hiragana/Katakana N5"
}
Field Default Description
primary_key First key of first item Field used to track progress
num_items 10 Number of items per session
subject_name (none) Display name for the subject

Commands

# Get new items (automatically prevents duplicates)
python3 ${OPENCLAW_SKILL_ROOT}/scripts/get_words.py

How to Handle Output

The output of get_words.py will be a raw list of items with all their fields. When you receive the output:

  1. Process the items exactly as instructed by the user's prompt or the automated Cron job orchestrating this skill.
  2. OPTIONAL: If you have access to the quizbuild tool, you can generate a short practice quiz based on the newly retrieved study items. When calling quizbuild__auto_create_exam, you follow this parameter structure:
{
  "title": "Quick Review Practice Exam",
  "questions": [
    {
      "content": "Question goes here?",
      "type": "multiple_choice",
      "answers": [
        {"content": "Answer 1", "correct": true},
        {"content": "Answer 2", "correct": false}
      ]
    }
  ]
}

If you do not have quizbuild, simply list out the practice items and act as a friendly tutor!

File Structure

File Purpose
data/data.json Study data (required)
data/config.json Configuration (optional)
data/learned_items.json Progress tracking (auto-generated)
scripts/get_words.py Main script
references/EXAMPLES.md Data formatting examples