Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Law Exam Trainer

v1.0.0

法考复习助手 - 根据法考视频/文档资料整理题库,支持选择题练习、错题解析、知识点定位。当用户说:法考、刷题、司法考试、法律职业资格考试、练习题、错题整理时触发。

0· 102·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for lujun2508/law-exam-trainer.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Law Exam Trainer" (lujun2508/law-exam-trainer) from ClawHub.
Skill page: https://clawhub.ai/lujun2508/law-exam-trainer
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install law-exam-trainer

ClawHub CLI

Package manager switcher

npx clawhub@latest install law-exam-trainer
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The name/description match the instructions (extract text from videos/docs, generate questions, provide explanations). However the SKILL.md assumes use of Whisper/OCR and other tooling without declaring dependencies or how those tools are invoked, and references third-party repos as resources rather than integrated libraries. This is plausible for the purpose but under-specified.
!
Instruction Scope
Instructions ask the agent to extract audio (Whisper), run OCR, and store a question bank, but give no concrete safe boundaries or commands. They reference uploading materials "给 CEO" and QQ dialogue as the UI, which is ambiguous. The SKILL.md also specifies a hard-coded local path (C:\Users\lujun\openclaw-projects\law-exam\题库\) for storing data — that indicates filesystem access expectations that are not declared and may not be appropriate or portable.
Install Mechanism
This is an instruction-only skill with no install spec or code to fetch — lowest install risk. That said, because it references Whisper/OCR, the agent may attempt to call or install those tools at runtime; the lack of declared dependencies leaves that behavior unspecified.
Credentials
No environment variables or credentials are requested (good). But the presence of a specific user-local file path suggests the skill expects read/write access to the host filesystem; the path is hard-coded to a specific user account (lujun), which is inconsistent and may cause the agent to access unexpected locations if followed literally.
Persistence & Privilege
always is false and there are no install hooks or claims to modify other skills or system settings. The skill does instruct storing files, but does not request persistent platform privileges in metadata.
What to consider before installing
This skill appears to do what it says, but it is under-specified and contains a hard-coded local path. Before installing or using it: (1) Do not upload sensitive or confidential documents until you know where transcription/OCR runs (local vs remote) and whether external APIs or services are used. (2) Ask the author how Whisper/OCR and OCR are invoked and whether any binaries will be installed or external APIs called (and if so, which credentials are required). (3) Confirm where question banks are saved and change the hard-coded path to a safe, user-controlled location. (4) Test with dummy data first. (5) If you need stronger assurance, request a version that declares dependencies and exact runtime commands or provide your own controlled pipeline for transcription/OCR so the skill doesn't run arbitrary installs or access unexpected filesystem locations.

Like a lobster shell, security has layers — review code before you run it.

latestvk972g5n5n64rk3016cxzndcpsh858dm1
102downloads
0stars
1versions
Updated 1w ago
v1.0.0
MIT-0

法考复习助手 Skill

功能概览

  1. 内容输入 — 上传法考视频/文档 → AI 提取内容
  2. 题目生成 — 从资料中出选择题/判断题题库
  3. 练习模式 — 做答 → 即时反馈对错
  4. 错题解析 — 答错 → 告诉哪里错 + 涉及知识点

题库来源

类型说明
视频文件上传视频 → 提取音频 → Whisper 转文字 → 出题
文字文档直接上传/粘贴 → 提取关键知识点 → 出题
图片/讲义OCR 识别 → 整理知识点 → 出题

题目类型

  • 单选题 / 多选题 / 判断题
  • 答案立即反馈
  • 解析包含:正确答案 + 错因 + 涉及法条

相关资源

  • GitHub: fighting41love/funNLP(含法律词库/罪名法务名词)
  • ClawHub: legal-document-assistantchina-legal-analysis

使用方式

  1. 发资料 — 上传视频/文档给 CEO
  2. 出题 — CEO 生成题库
  3. 练习 — QQ 对话形式做答,CEO 评判对错

题库目录

题库文件保存在:C:\Users\lujun\openclaw-projects\law-exam\题库\

Comments

Loading comments...