Skill flagged — suspicious patterns detected

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

瑞幸点单MCP

v0.0.1

瑞幸咖啡订单 MCP 服务。提供一键下单、门店查询、商品浏览、购物车管理、订单管理、优惠券/咖啡券管理、地址管理等功能。适用于瑞幸咖啡内部订购系统。

0· 67·0 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 fjmjulzl/lk-order-mcp.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "瑞幸点单MCP" (fjmjulzl/lk-order-mcp) from ClawHub.
Skill page: https://clawhub.ai/fjmjulzl/lk-order-mcp
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 lk-order-mcp

ClawHub CLI

Package manager switcher

npx clawhub@latest install lk-order-mcp
Security Scan
Capability signals
CryptoCan make purchasesRequires OAuth tokenRequires sensitive credentials
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
high confidence
Purpose & Capability
The skill's name, description, SKILL.md, README and code all describe a Luckin (瑞幸) MCP ordering client and the implemented tools match that purpose. However the registry metadata claims 'Required env vars: none' and 'Primary credential: none' while clawhub.json, README, SKILL.md and the code expect a single sensitive credential (LK_ORDER_TOKEN / Bearer token). This is an internal inconsistency: the token is legitimately required for the stated purpose but the registry metadata fails to declare it.
!
Instruction Scope
Runtime instructions and code direct the agent to make network calls to the stated MCP endpoint and to persist a session file at /tmp/lkorder-session-id.txt — which is consistent with the service. However the SKILL.md and README include example curl commands that embed a long Bearer token directly (hard-coded token in examples). The code (lk-order.cjs) executes curl via child_process.execSync and constructs command-line strings that include the token, which exposes the token to process listings and shell histories. The skill also downloads QR images (executes curl to arbitrary qrImageUrl returned by server) and writes files to /tmp; these behaviors are within ordering functionality but increase the attack surface and risk of credential/process exposure.
Install Mechanism
There is no remote download/install spec in the registry (instruction-only), and the included install.sh copies local files into the OpenClaw workspace — no external fetch during install. That is lower risk than remote installers. Minor concern: the registry said 'no install spec' yet an install.sh and executable node script are included (not a security problem but a metadata mismatch).
!
Credentials
The skill legitimately needs one sensitive credential (LK_ORDER_TOKEN / Bearer token) to call the MCP endpoint — that is proportional to its purpose. Problems: the public registry metadata did not declare this requirement, and SKILL.md/README contain an example with a hard-coded Bearer token (a1f6cb06...), which appears to be a leaked credential in the documentation. The code will also look for openclaw.json in multiple locations (including ~/.openclaw/openclaw.json and /home/node/.openclaw/openclaw.json), meaning it will read local configuration files; this is expected but worth noting because misconfigured files may contain other secrets.
Persistence & Privilege
The skill does not request 'always: true', does not modify other skills, and its install script writes only to the skill workspace and creates /tmp files for session and downloaded QR images. Temporary persistence in /tmp and storing a session ID is normal for this use case. No evidence it tries to persist credentials or alter system-wide settings beyond reading/writing OpenClaw config files and its own skill directory.
What to consider before installing
This package implements the described ordering features, but take these precautions before installing: - Expect to provide a sensitive token (LK_ORDER_TOKEN). The registry metadata failed to declare it — do not assume none is needed. Configure the token via environment variable or ~/.openclaw/openclaw.json, not by copying tokens into examples or files you share. - The SKILL.md/README include a hard-coded Bearer token in curl examples. Treat that token as leaked: do not reuse it. If you see that token active in your environment, rotate/reset it immediately. - The Node code uses child_process.execSync to run curl with the token included on the command line. This exposes the token to other local users via process listings and shell history. Prefer running the skill in a trusted, single-user environment or modify the code to use a library HTTP client (which keeps credentials out of process args) before use. - The skill will read ~/.openclaw/openclaw.json and other local paths for configuration; review those files to ensure they do not contain unrelated secrets you don't want accessed. - Verify the MCP endpoint domain (inpre.lkcoffee.com / openpre.lkcoffee.com) and confirm the skill's provenance. The package owner and homepage fields are generic; if you do not trust the source, avoid installing. If you proceed, remove hard-coded tokens from documentation, run the skill with the token set only as an environment variable (not passed on the command line), and rotate the token if it was ever exposed.
lk-order.cjs:100
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.

latestvk97f5qx02tdb3yt6vwzg90a7as84zpm3
67downloads
0stars
1versions
Updated 1w ago
v0.0.1
MIT-0

lk-order MCP Skill

瑞幸咖啡内部订单系统 MCP 服务封装。支持一键下单门店查询商品浏览购物车管理订单管理优惠券/咖啡券地址管理等完整功能。

触发条件

当用户提到以下关键词时触发本 Skill:

☕ 品牌/场景触发(最高优先级)

  • 品牌词: "瑞幸"、"luckin"、"瑞幸咖啡"
  • 需求词: "饿"、"饿了吗"、"想吃"、"想吃东西"、"肚子饿"
  • 吃喝词: "喝"、"喝的"、"吃"、"吃的"、"来杯"、"来份"
  • 品类词: "饮品"、"饮料"、"轻食"、"面包"、"蛋糕"、"三明治"、"小吃"

🛒 下单相关

  • 动作词: "下单"、"点咖啡"、"买咖啡"、"订购"、"点单"、"点餐"、"买点"、"整一杯"
  • 商品名: "特仑苏"、"美式"、"拿铁"、"生椰"、"瑞纳冰"、"燕麦"、"摩卡"、"卡布奇诺"

🏪 门店查询

  • "附近门店"、"门店列表"、"店铺"、"瑞幸店"、"哪里有店"、"最近门店"

📋 订单管理

  • "我的订单"、"订单详情"、"取消订单"、"删除订单"、"查订单"、"订单状态"

🛒 购物车

  • "购物车"、"加入购物车"、"清空购物车"、"看看购物车"

🎫 优惠券/卡券

  • "优惠券"、"咖啡券"、"咖啡钱包"、"咖啡库券"、"有什么券"、"用券"

📍 地址管理

  • "收货地址"、"添加地址"、"修改地址"、"外送地址"

触发示例

自动触发场景

  • "瑞幸有什么新品" → 触发技能,查询菜单
  • "有点饿,想吃的" → 触发技能,推荐轻食/饮品
  • "渴了,想喝点东西" → 触发技能,推荐饮品
  • "来杯喝的" → 触发技能,询问具体需求
  • "瑞幸,下单美式" → 直接下单
  • "附近有没有瑞幸" → 查询附近门店

核心能力

1. 一键下单(最常用)

用户只需说想喝什么,自动完成全流程。

示例:

  • "帮我下单美式"
  • "点一杯热拿铁,大杯"
  • "下单特仑苏"

2. 门店查询

根据经纬度获取附近瑞幸门店。

3. 商品浏览

获取菜单、商品详情、价格计算。

4. 购物车管理

添加商品、查看购物车、结算。

5. 订单管理

查看订单列表、订单详情、取消/删除订单。

6. 优惠券/卡券

获取可用优惠券、咖啡钱包、咖啡库券。

7. 地址管理

添加/修改/删除收货地址(外送用)。


MCP 服务配置

配置项
服务 URLhttps://inpre.lkcoffee.com/app/proxymcp
认证方式Bearer Token
协议版本2024-11-05
Session 文件/tmp/lkorder-session-id.txt

可用工具(35 个)

🚀 下单相关(6 个)

工具名说明必填参数
quick_order一键下单(自动找门店→选商品→创建订单→支付)keyword(想喝什么)
create_order创建订单deptId, delivery, productList
preview_order预览订单(计算价格、优惠、配送费)deptId, delivery, productList
pay_order支付订单orderId
cancel_order取消订单orderId
delete_order删除订单orderList

🏪 门店/商品(5 个)

工具名说明必填参数
get_nearby_shops获取附近门店列表无(自动根据 IP 定位)
get_shop_detail获取门店详情deptId
get_menu获取门店菜单deptId
get_product_detail获取商品详情productId, deptId
calculate_price计算商品价格productId, deptId, skuCode

🛒 购物车(5 个)

工具名说明必填参数
get_shopping_cart获取购物车信息deptId
add_to_cart加入购物车productId, skuCode
clear_cart清空购物车
set_cart_checked设置购物车行勾选状态skuCodes, checked
checkout_cart结算购物车deptId, delivery

🎫 优惠券/卡券(4 个)

工具名说明必填参数
get_coupon_list获取我的优惠券列表
get_ticket_list获取下单时可用的优惠券deptId, productList
get_coffee_wallet_list获取咖啡钱包列表
get_coffee_store_list获取下单时可用的咖啡库券deptId, productList

📍 地址管理(5 个)

工具名说明必填参数
get_user_addresses获取收货地址列表
search_address搜索地址keyName
add_address添加收货地址userName, sex, tel, address
update_address更新收货地址addrId, userName, sex, tel, address
delete_address删除收货地址addrId

📋 订单管理(3 个)

工具名说明必填参数
get_order_list获取订单列表
get_order_detail获取订单详情orderId
get_pay_list获取支付方式列表deptId

⚙️ 其他(7 个)

工具名说明必填参数
session_info查看登录状态
logout退出登录
restore_session恢复会话token
get_activity_list获取营销活动优惠列表deptId, productList
get_remark_options获取订单备注选项deptId, productList
get_city_list获取已开通城市列表
get_user_city根据经纬度定位城市longitude, latitude

使用示例

一键下单(推荐)

# 调用脚本
/home/node/.openclaw/scripts/lkorder-mcp.sh call quick_order '{"keyword":"美式"}'

# 参数说明
{
  "keyword": "美式",           # 必填:想喝什么
  "temperature": "热",         # 可选:温度偏好(热/冰)
  "cup": "大杯",               # 可选:杯型
  "sugar": "标准糖",           # 可选:甜度
  "amount": 1,                 # 可选:数量,默认 1
  "payType": "1",              # 可选:支付方式(1=支付宝,2=微信,7=余额)
  "useCafeKu": true,           # 可选:是否使用咖啡券,默认 true
  "useDiscount": true          # 可选:是否使用优惠券,默认 true
}

获取附近门店

/home/node/.openclaw/scripts/lkorder-mcp.sh call get_nearby_shops '{"latitude":24.4798,"longitude":118.0894}'

查看购物车

/home/node/.openclaw/scripts/lkorder-mcp.sh call get_shopping_cart '{"deptId":"12345"}'

获取优惠券列表

/home/node/.openclaw/scripts/lkorder-mcp.sh call get_coupon_list '{}'

查看订单列表

/home/node/.openclaw/scripts/lkorder-mcp.sh call get_order_list '{"pageNo":1}'

调用方式

方式 1:使用 Shell 脚本(推荐)

# 初始化会话(首次使用或 session 过期时)
/home/node/.openclaw/scripts/lkorder-mcp.sh init

# 调用工具
/home/node/.openclaw/scripts/lkorder-mcp.sh call <方法名> '[参数 JSON]'

# 示例
/home/node/.openclaw/scripts/lkorder-mcp.sh call session_info
/home/node/.openclaw/scripts/lkorder-mcp.sh call quick_order '{"keyword":"拿铁"}'

方式 2:直接 curl 调用

# 1. 初始化获取 Session ID
SESSION_ID=$(curl -s -i -X POST \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "Authorization: Bearer a1f6cb06a75d27f3e822c64077fd46acd33cde6b36b30ff9bec468a2ef78f136" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{}}' \
  https://inpre.lkcoffee.com/app/proxymcp | grep -i "^mcp-session-id:" | awk '{print $2}')

# 2. 调用工具
curl -s -X POST \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "Authorization: Bearer a1f6cb06a75d27f3e822c64077fd46acd33cde6b36b30ff9bec468a2ef78f136" \
  -H "mcp-session-id: ${SESSION_ID}" \
  -d '{"jsonrpc":"2.0","id":2,"method":"session_info"}' \
  https://inpre.lkcoffee.com/app/proxymcp

会话管理

Session ID 持久化

  • 保存位置: /tmp/lkorder-session-id.txt
  • 有效期: 由服务端决定(通常数小时)
  • 自动恢复: 脚本会自动读取已保存的 Session ID

重置会话

/home/node/.openclaw/scripts/lkorder-mcp.sh reset

错误处理

错误信息原因解决方案
缺少有效的会话 IDSession 过期或丢失执行 init 重新初始化
请求错误:参数校验失败参数格式错误检查必填参数和 JSON 格式
未找到门店经纬度无效或附近无门店检查经纬度或扩大搜索范围
商品已售罄商品库存不足更换商品或选择其他门店
支付方式不可用该门店不支持指定支付方式更换支付方式

注意事项

  1. 首次使用必须先初始化:执行 init 获取 Session ID
  2. Session ID 必须携带:所有后续请求需包含 mcp-session-id header
  3. 参数格式严格:JSON 参数必须符合 schema 定义
  4. 坐标系统:经纬度使用 GCJ-02 标准(国内)
  5. 支付方式
    • 1 = 支付宝(返回支付链接)
    • 2 = 微信(返回支付链接)
    • 7 = 余额(直接扣款,推荐)
  6. 取餐方式
    • pick = 自取
    • delivery = 外送(需填写地址)

快速参考

最常用场景

# 1. 一键下单(最简单)
./lkorder-mcp.sh call quick_order '{"keyword":"美式"}'

# 2. 查看登录状态
./lkorder-mcp.sh call session_info

# 3. 获取附近门店
./lkorder-mcp.sh call get_nearby_shops '{"latitude":24.4798,"longitude":118.0894}'

# 4. 查看我的订单
./lkorder-mcp.sh call get_order_list '{"pageNo":1}'

# 5. 获取优惠券
./lkorder-mcp.sh call get_coupon_list '{}'

# 6. 查看购物车
./lkorder-mcp.sh call get_shopping_cart '{"deptId":"12345"}'

相关文件

文件说明
/home/node/.openclaw/scripts/lkorder-mcp.shMCP 调用脚本
/tmp/lkorder-session-id.txtSession ID 保存文件
/home/node/.openclaw/openclaw.jsonOpenClaw 配置(含 Token)

安全提醒

⚠️ Bearer Token 属于敏感凭证,请勿:

  • 提交到代码仓库
  • 发送给无关人员
  • 在公开场合展示

如需更新 Token,请编辑 /home/node/.openclaw/openclaw.json 中的 channels.wecom.mcpServers.lk-order.headers.Authorization 字段。

Comments

Loading comments...