← 返回 Skills 市场

TOEFL Speaking Question Generator

Name: TOEFL Speaking Question Generator
Author: saikewei

作者 saikewei · GitHub ↗ · v1.0.1 · MIT-0

cross-platform ✓ 安全检测通过

总下载

当前安装

版本数

在 OpenClaw 中安装

/install toefl-speaking

功能描述

托福口语面试部分出题引擎。AI 自动选定主题，生成 4 道递进式面试题和参考答案。不依赖固定主题库。

使用说明 (SKILL.md)

TOEFL Speaking 出题引擎

核心能力

输入一个主题 → 输出 4 道递进式问题 + 隐藏的参考答案（存入文件，不展示）。

出题规则

主题选择

主题由 AI 完全自主决定。每次出题时：

读取 toefl/history.json，查看最近 10 个已出过的主题
从中等或高等难度的学术/校园话题中选一个与之不同的方向
尽可能保证与最近 10 次不重复、话题跨度大、有随机性

题目结构

Q1 (Past/Experience) → Q2 (Habit/Preference) → Q3 (Deep Opinion) → Q4 (Future/Policy)

参考文献

出题格式和答案标准见 prompt-template.md。

与用户互动

用户说	我做
"来一套题" / "出题"	1. 读取 `toefl/history.json` 看最近10个主题
2. 选新主题出4道题，生成参考答案
3. 写入 `toefl/latest.md`（含完整题目+答案+解析）
4. 更新 `toefl/history.json`（追加新记录）
5. 复制至 `toefl/YYYY-MM-DD.md` 存档
6. 只向用户展示4道问题，隐藏答案
"今天的答案" / "Q1的答案"	读 `toefl/latest.md` 展示对应答案+解析
"评价一下我的回答"	用户给出回答后，按评分标准反馈（流利度、词汇、语法、内容）
"换一个" / "换个话题"	重选主题重新出题，标记旧题作废

数据存储

toefl/
├── latest.md          # 最新一套题（含答案，供查看答案时读取）
├── history.json       # 出题记录（主题+日期+是否看过答案）
└── YYYY-MM-DD.md     # 每日存档

注意事项

不依赖固定主题列表，AI 自主生成主题
不包含任何定时/调度逻辑，触发方式由外部决定

安全使用建议

This skill appears safe for ordinary TOEFL practice use. Before installing, be comfortable with it creating local files under toefl/, and remember that its generated questions and answers are practice material rather than verified official TOEFL content.

功能分析

Type: OpenClaw Skill Name: toefl-speaking Version: 1.0.1 The skill bundle is a TOEFL speaking practice engine that generates interview questions and stores them in a local directory. It uses standard file operations for history tracking (toefl/history.json) and prompt templates for content generation (prompt-template.md). No evidence of malicious intent, data exfiltration, or unauthorized command execution was found in SKILL.md or the associated files.

能力评估

ℹ Purpose & Capability

The TOEFL question-generation workflow is coherent and purpose-aligned; note that prompt-template.md frames the model as an “官方口语教研员,” which may overstate official authority.

✓ Instruction Scope

The skill is user-invoked, has no scheduler, and the autonomous topic choice is limited to generating practice questions.

ℹ Install Mechanism

There is no install code or dependency execution, but the registry version is 1.0.1 while SKILL.md declares version 2.1.0, creating a minor provenance/packaging inconsistency.

ℹ Credentials

The skill reads and writes only scoped local study files under the toefl/ directory, which is proportionate to its practice-history purpose.

ℹ Persistence & Privilege

It persists latest questions, answers, topic history, and daily archives locally, but requests no credentials, privileged access, or background persistence.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install toefl-speaking
安装完成后，直接呼叫该 Skill 的名称或使用 /toefl-speaking 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.1

- Enhanced the user interaction section to clarify step-by-step output, answer storage, and file update workflows. - Updated the reference to the answer template file from prompts/prompt-template.md to prompt-template.md. - No code changes; documentation refinement only.

v1.0.0

Version 2.0.0 introduces a fully autonomous TOEFL speaking exam question engine. - AI now independently generates themes, ensuring each set is unique and not from a fixed pool. - Questions escalate in depth: experience → habit/preference → deep opinion → future perspective. - Maintains a history to avoid recent topic repetition and maximize variety. - Reference answers are stored but kept hidden during initial question delivery. - New structured file system for question sets, history, and daily archives. - Interactions include question delivery, answer review, user answer evaluation, and theme renewal.

元数据

Slug toefl-speaking

版本 1.0.1

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 2

常见问题