Skill flagged — suspicious patterns detected

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

中医执业医师备考助手

v1.0.0

中医执业医师考试备考助手 - 生成每日复习计划、评估答案、管理错题集

0· 120·0 current·0 all-time
byflyingants@tom859174-sketch

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for tom859174-sketch/tcm-exam-assistant.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "中医执业医师备考助手" (tom859174-sketch/tcm-exam-assistant) from ClawHub.
Skill page: https://clawhub.ai/tom859174-sketch/tcm-exam-assistant
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Required binaries: node
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 tcm-exam-assistant

ClawHub CLI

Package manager switcher

npx clawhub@latest install tcm-exam-assistant
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description (exam plan, answer evaluation, wrong-question management) match the code and SKILL.md. Required runtime (node) is appropriate and the skill only declares local storage under a memory/ directory.
Instruction Scope
SKILL.md and assistant.js keep scope to plan generation, answer evaluation and local wrong-question storage. The skill references syncing documents from the user's Desktop and returning an instruction object for the OpenClaw host to read docs; the sync function is currently a no-op stub. This means the skill may expect the host to supply document content; it does not currently disclose reading arbitrary unrelated files, but the Desktop-sync mention is a place to double-check before enabling.
Install Mechanism
No install spec; skill is a single Node script and requires only node on PATH. Nothing is downloaded or installed by the skill itself.
Credentials
No environment variables, external API keys, or config paths are requested. Memory files are created under the skill's own memory/ directory inside the skill folder.
Persistence & Privilege
always:false and the skill persists only to its own memory files (stage.json, progress.json, wrong-questions.json, etc.). It does not request system-wide configuration changes. This is a normal level of persistence for a study-assistant skill.
Assessment
This skill appears to do what it says: generate plans, grade answers, and keep a local wrong-question database. Before installing or enabling it: 1) Review the full assistant.js (the provided preview was truncated) to confirm there are no network calls, hidden endpoints, or uses of child_process.execSync with untrusted input — the script imports execSync which can run shell commands if used later. 2) Note the skill writes files under its own memory/ directory; if you have sensitive files, do not point any sync functionality at Desktop or other private folders until you inspect that code. 3) There is a robustness bug in the visible code (OUTPUT_DIR is referenced but not defined) — expect potential runtime errors. 4) If you will allow autonomous invocation, consider running it in a restricted environment or sandbox first and back up any important local data. If you want higher assurance, provide the remaining lines of assistant.js so they can be reviewed for any use of execSync, network requests, or filesystem operations outside the skill directory.
assistant.js:228
Shell command execution detected (child_process).
Patterns worth reviewing
These patterns may indicate risky behavior. Check the VirusTotal and OpenClaw results above for context-aware analysis before installing.

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

Runtime requirements

📚 Clawdis
Binsnode
acupuncturevk97fvbrgtgk9m6def68p19s2gh83aev0chinese-medicinevk97fvbrgtgk9m6def68p19s2gh83aev0exam-prepvk97fvbrgtgk9m6def68p19s2gh83aev0latestvk97fvbrgtgk9m6def68p19s2gh83aev0openclaw-skillvk97fvbrgtgk9m6def68p19s2gh83aev0study-assistantvk97fvbrgtgk9m6def68p19s2gh83aev0tcmvk97fvbrgtgk9m6def68p19s2gh83aev0
120downloads
0stars
1versions
Updated 1mo ago
v1.0.0
MIT-0

tcm-exam-assistant 技能 v2.0

功能描述

中医执业医师考试备考助手,专注于复习计划生成、答案评估和错题管理

注意:文档读取(docx/txt/pdf/表格)由 OpenClaw 主程序/AI 处理,本 skill 通过 simple-memory 持久化存储学习进度。

核心功能

  1. 每日计划生成 - 根据用户反馈生成次日复习计划(新学 + 复习)
  2. 答案评估 - 批改用户提交的真题答案,指出错误点并解析
  3. 错题管理 - 记录错题到错题集,支持导出打印
  4. 进度追踪 - 记录学习进度,支持笔试→技能考试阶段切换
  5. 激进模式 - 支持 conservative/standard/aggressive/crazy 四种学习强度

多 Skills 协作架构

┌─────────────────────────────────────────────────────────┐
│                    OpenClaw 主程序                       │
│  (AI 理解用户意图,调度各个 skills)                      │
└─────────────────────────────────────────────────────────┘
                            │
        ┌───────────────────┼───────────────────┐
        │                   │                   │
        ▼                   ▼                   ▼
┌───────────────┐  ┌───────────────┐  ┌───────────────┐
│ OpenClaw AI   │  │ tcm-exam-     │  │ save-to-word  │
│ 文档读取       │  │ assistant     │  │ 导出错题集    │
│ - docx        │  │ - 生成计划     │  │ - docx        │
│ - txt         │  │ - 评估答案     │  │               │
│ - pdf         │  │ - 错题管理     │  │               │
│ - 表格         │  │ - 进度追踪     │  │               │
└───────────────┘  └───────────────┘  └───────────────┘
        │                   │                   │
        └───────────────────┼───────────────────┘
                            ▼
                  ┌─────────────────┐
                  │  simple-memory  │
                  │  (持久化存储)    │
                  └─────────────────┘

使用场景

场景 1:上传备考文档,生成复习计划

用户:上传 2026 执考 -85-中医执考复习大提点.docx
      根据文档内容,生成我的复习计划

流程

  1. OpenClaw AI 读取 docx 文档
  2. 调用 tcm-exam-assistant 生成计划
  3. 输出:每日复习计划

场景 2:动态提问

用户:我目前在中内第三章,昨晚咳嗽没看懂,今天应该看什么?

流程

  1. OpenClaw AI 理解问题
  2. 调用 tcm-exam-assistant 生成针对性计划
  3. 输出:今日任务

场景 3:提交答案,评估错题

用户:今天做了 10 道题,答案:1.A 2.B 3.C 4.A 5.B 6.C 7.A 8.B 9.C 10.A
      正确答案:1.A 2.B 3.A 4.A 5.B 6.C 7.A 8.A 9.C 10.A

流程

  1. 调用 tcm-exam-assistant 评估答案
  2. 记录错题到 simple-memory
  3. 输出:正确率 + 错题分析

场景 4:导出错题集

用户:导出我的错题集为 Word 文档

流程

  1. 调用 tcm-exam-assistant 获取错题数据
  2. 调用 save-to-word 生成 docx 文件
  3. 输出:Word 文档路径

调用方式

1. 生成每日计划

node <skill-dir>/assistant.js "generate-plan" "{{用户反馈}}"

用户反馈示例

今天复习了技能第一站,病例分析还不熟
今天复习了中内肺系病证第 3-4 节(感冒、咳嗽),中午背了麻黄桂枝对比,晚饭后看了方剂学麻黄汤,做了 10 道真题,错了 3 道(第 2、5、8 题)

2. 评估答案

node "C:\Users\梁\.openclaw\skills\tcm-exam-assistant\assistant.js" "evaluate-answers" "{{题目}}" "{{用户答案}}" "{{正确答案}}"

3. 导出错题集

node "C:\Users\梁\.openclaw\skills\tcm-exam-assistant\assistant.js" "export-wrong-questions"

核心特性

阶段自动判断

  • 技能考试阶段(考前 90 天内):以技能三站训练为主
  • 笔试阶段:以中医内科学、中药、方剂等笔试科目为主

时间适配

  • 中午 30 分钟:只安排背诵内容(口诀、表格、对比)
  • 晚上 3 小时:安排理解、做题、错题分析

激进模式

模式系数适用场景
conservative1.0基础薄弱、时间充裕
standard1.2正常复习节奏
aggressive1.5时间紧张、冲刺阶段(默认)
crazy2.0极限冲刺、考前 30 天

记忆文件

文件说明
memory/stage.json当前备考阶段(skill-exam-prep / written-exam-prep)
memory/progress.json学习进度(按科目)
memory/wrong-questions.json错题集
memory/review-schedule.json复习调度(艾宾浩斯曲线)
memory/user-profile.json用户画像(偏好/习惯)
memory/doc-cache.json文档缓存(考试信息)

科目列表

科目代码科目名称
skill-station-1技能第一站(病例分析)
skill-station-2技能第二站(操作)
skill-station-3技能第三站(答辩)
tcm-herbs中药学
tcm-formulas方剂学
tcm-internal中医内科学
acupuncture针灸学
tcm-basic中医基础理论
tcm-diagnosis中医诊断学
western-medicine西医综合

输出路径

  • 错题集导出:用户配置的输出目录
  • 文档目录:用户配置的备考文档目录

考试日期配置

assistant.js 中修改:

const SKILL_EXAM_DATE = new Date('2026-06-06');   // 技能考试
const WRITTEN_EXAM_DATE = new Date('2026-08-15'); // 笔试

Comments

Loading comments...