Skill flagged — suspicious patterns detected

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

mt-travel-ai

v1.0.2

基于美团酒旅供给,处理旅游出行需求,包括提供酒店、机火、门票、度假等商品的查询交易能力,以及定制化旅行攻略能力,打通从“灵感启发”到“一键下单”的全链路。

0· 131·0 current·0 all-time
byFitzwilliam Zhang@qy-zhang

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for qy-zhang/mt-travel-ai.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "mt-travel-ai" (qy-zhang/mt-travel-ai) from ClawHub.
Skill page: https://clawhub.ai/qy-zhang/mt-travel-ai
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 mt-travel-ai

ClawHub CLI

Package manager switcher

npx clawhub@latest install mt-travel-ai
Security Scan
Capability signals
Requires OAuth token
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The skill's description (Meituan travel assistant) matches the runtime behavior: it uses a travel CLI and an API token to query Meituan data. However, the manifest declares no required binaries or env vars while SKILL.md requires installing and invoking an npm CLI (mttravel) and reading/writing ~/.config/meituan-travel/config.json — this mismatch between manifest and runtime instructions is an inconsistency to be aware of.
!
Instruction Scope
SKILL.md explicitly instructs the agent to prompt the user to paste an API token into the chat, then run shell commands to create ~/.config/meituan-travel/config.json containing that token. It also mandates absolute passthrough of CLI output (images/links) and special behavior for WeChat (call a 'message' tool). These instructions cause the agent to read/write the user's home filesystem and to relay whatever the CLI returns without filtering — increasing risk of accidental disclosure or transmission of sensitive or malicious content. There is also a contradictory emphasis ('Token is highly sensitive, do not print token in chat') paired with asking the user to paste the token into the conversation.
Install Mechanism
There is no install spec in the registry metadata, but SKILL.md instructs users/agents to run 'npm i -g @meituan-travel/travel-cli' and then use 'mttravel'. Requiring a global npm package is a moderate-risk install vector (traceable to npm package metadata if you verify it) and should be validated by the user. The manifest not listing this requirement is another inconsistency.
!
Credentials
The skill does not request environment variables, but it requires an API token (stored at ~/.config/meituan-travel/config.json). Having users paste tokens into chat and saving them to a plaintext file is disproportionate from a security perspective: the token could be exfiltrated via logs, conversation history, or if the agent relays CLI output. The token-handling workflow lacks guidance for scoping/revoking tokens or restricting file permissions.
!
Persistence & Privilege
The skill writes a persistent plaintext credential file in the user's home directory (~/.config/meituan-travel/config.json). While always:false (not force-included), this persistent storage of sensitive credentials in an obvious path is a privilege/persistence risk — especially on shared machines or if backups/agent logs include that file. The skill does not instruct encrypting the token or tightening file permissions.
What to consider before installing
This skill appears to do what it claims (call a Meituan travel CLI) but has multiple security concerns you should weigh before installing or using it. Key points to consider: - Do not paste highly sensitive tokens into chat unless you fully trust the recipient; prefer creating a limited-scope token or a separate test account for this skill. - The skill instructs the agent to save your token in plaintext at ~/.config/meituan-travel/config.json. If you proceed, inspect that file, set strict permissions (chmod 600), and consider deleting the token when finished. - Verify the npm package (@meituan-travel/travel-cli) before installing: check publisher identity and package source on the npm registry. Prefer installing manually yourself rather than having the agent run installs. - Be cautious about the mandated 'passthrough' of CLI output (images/links): it may relay unexpected or sensitive content verbatim. If you use this skill in a channel you don't fully control, avoid sending secrets there. - Ask the skill author (or maintainer) to fix inconsistencies: declare required binaries/env in the manifest, remove the requirement to paste tokens into chat (support an out-of-band secure token entry), and recommend secure storage (encrypted or OS keyring) rather than plaintext files. If you need higher assurance, request an implementation or package audit (package source, publisher identity, and a review of the CLI) before providing real credentials.

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

latestvk97ak22q7c2rbxvhy503rc9esx84kmcg
131downloads
0stars
3versions
Updated 2w ago
v1.0.2
MIT-0

美团旅行助手 Skill

为用户提供完整的旅行服务,包括目的地推荐、行程规划、酒店推荐预订、机票/火车票查询、景点门票、度假跟团等。

适用场景边界

使用此 skill:

  • "想去踏青赏花,推荐几个必去的城市"
  • "周末两天适合去哪里玩"
  • "带小孩去哪里旅游比较好"
  • "春天适合去的国内景点"
  • "从北京出发去哪玩"
  • "明天去武汉的火车票"
  • "去南方的特价机票"
  • "北京到清远的交通方案"
  • "两大一小怎么买上海迪士尼门票"
  • "帮我订这周末开封的情侣酒店,预算500内离万岁山1公里内"

不使用此 skill:

  • 出国签证申请、护照办理流程
  • 非旅行相关的外卖、打车、跑腿等美团其他业务

核心执行流程

  1. 提取参数 — 识别用户的「当前定位城市」(获取不到默认北京)和「查询需求」。若用户明确指定了出发地,以用户指定为准。
  2. 鉴权准备(Token 检查) — 调用 CLI 前,必须先执行下方的【Token 管理流程】。
  3. 安抚等待 — 该 API 执行耗时较长(约 1-2 分钟),请务必先向用户发送:

    🔍 正在连接美团酒旅数据接口为您规划,耗时约 1-2 分钟,请稍候...

  4. 执行 CLI — 使用 mttravel 调用 API,传入参数。
  5. 解析与渲染输出 — 严格按照下方的【输出规范】向用户展示最终结果。

Token 管理流程

Step 1:检查本地配置

检查 ~/.config/meituan-travel/config.json 是否存在且包含 key 字段。

  • ✅ 如果存在且非空 → 读取 token,直接进入 Step 4
  • ❌ 如果不存在或为空 → 进入 Step 2

Step 2:引导用户创建

向用户发送以下原话(务必原样输出):

🔑 需要配置 API Token 请前往美团开发者中心入驻个人开发者并创建 Token: 点击此处前往创建 Token 创建完成后,请将 Token 字符串发给我,我会帮您保存并继续查询。

(等待用户回复 Token)

Step 3:保存 Token

收到 Token 后,执行以下命令将其保存至 config.json

mkdir -p ~/.config/meituan-travel
cat > ~/.config/meituan-travel/config.json << 'EOF'
{
  "key": "<用户提供的token>"
}
EOF

保存成功后告知用户:「✅ Token 已保存,正在为您查询…」,然后继续执行。

Step 4 — 执行 CLI

CLI 会自动从 ~/.config/meituan-travel/config.json 读取 Token,直接执行: mttravel [城市] "<query>"


Token 失效处理(CLI 返回鉴权错误时)

当 CLI 执行返回错误信息包含以下关键词时,触发此流程:

  • "鉴权失败"
  • "无效的访问令牌"
  • "Token"
  • "未设置"
  • "access token"
  • "key"

必须立即执行以下步骤:

情况 A:config.json 存在但 Token 失效

向用户发送:

⚠️ Token 已失效,需要更新

请前往开发者中心重新创建 Token

⚠️ Token 为极高敏感凭证,禁止在对话中打印 Token 明文

创建完成后,请将新 Token 发送给我,我会更新并重新查询。

收到新 Token 后,执行:

cat > ~/.config/meituan-travel/config.json << 'EOF'
{
  "key": "<新token>"
}
EOF

然后重新执行 CLI 查询。

情况 B:config.json 不存在或为空

重新执行 Step 2 和 Step 3。


CLI 调用方式

前置要求

npm i -g @meituan-travel/travel-cli

执行命令

mttravel [城市] "<query>"

CLI 会自动从 ~/.config/meituan-travel/config.json 读取 Token。


⚠️ 严格输出规范

为保证数据真实性和信息完整性,在将工具结果返回给用户时,必须绝对遵守以下原则。无论用户使用的是微信、飞书还是任何其他客户端,你都必须无视渠道差异,严格执行以下标准 Markdown 输出,绝对禁止自作主张进行“格式降级”:

  1. 零压缩/零删减原则
  • 必须将 CLI 工具输出的全部文字内容原样呈现,绝不允许合并段落、删减字数。不得省略酒店名、商品名、价格、评分、标签、描述、图片、链接等转化关键信息。
  • 可做轻量排版整理,但不得删减事实信息,不得把完整描述压缩成只有几个要点。
  1. 渠道展示总原则
  • 渲染工作由最终客户端负责,大模型的职责仅限于输出完整的 Markdown 文本,绝不允许出现“吞掉图片”、“去除超链接只保留文字”的行为。
  • 接口返回的内容中若包含 Markdown 格式的图片(![alt](url))和超链接([text](url)),必须 100% 原样透传输出。
  • 严禁猜测或假设当前聊天渠道(如微信/飞书)不支持 Markdown 而擅自将图片和链接剥离。
  • 【微信渠道特供】:如果检测到当前用户处于 微信 渠道,你必须优先调用 message 工具,将包含完整 Markdown(图片、链接等)的内容作为参数传递给该工具发送。严禁在微信渠道直接将内容作为普通文本回复,否则会导致排版失效。
  1. 图片强制内嵌展示
  • CLI 输出中包含的图片格式(如 ![图片描述](图片URL)),必须原封不动输出。
  • 若 CLI 输出的是裸露的图片 URL(以 http 开头,包含 .jpg .png 等后缀),必须强制包裹为 Markdown 内嵌图片格式渲染:![](图片URL)
  • 红线规定:禁止仅展示图片链接文字而不渲染图片;禁止以“查看图片”超链接替代内嵌展示;图片必须紧跟对应商品内容之后,禁止移至末尾。
  1. 超链接强制保留:
  • 接口返回的跳转链接(如 [**隐山酒店**](http://...)),必须完整保留中括号和小括号的结构。
  • 若工具输出中有裸露的 URL 链接,必须转换为 Markdown 超链接格式,如:[点击查看详情](https://...)
  1. 真实数据合规强制标注
  • 凡涉及评分字段(如用户评分/口碑),必须强制修改为加粗的特定格式,例:4.8分(美团真实评分)
  • 凡涉及星级字段(如酒店星级),必须强制修改为,例:美团5星级
  • 距离等实时数据,在首次出现时需在句末或括号内补充标注 (美团实时数据)。
  • 除上述明确规定的「格式包装」,绝对禁止对工具返回的任何数值进行主观篡改、四舍五入或伪造。
  1. 价格原样输出
  • CLI 返回的价格字符串必须原样展示,禁止做任何解读、转换或补充说明。
  • 例如后端返回 ¥4XX起/晚,必须原样输出 ¥4XX起/晚,禁止转换为 ¥400+起/晚、¥400起/晚 等任何变体形式。
  • 价格中的占位符(如 X、XX、XXX)是后端的脱敏处理,不得自行还原或猜测实际数值。

🆘错误处理预案

遇到异常情况你的应对策略
网络超时(>120s)安抚用户:“请求超时啦,当前查询人数较多,请换个问法或稍后再试。”
查询失败展示错误信息,建议用户换个问法重试
城市无法识别停止猜测,主动询问用户确认具体的出发城市或目的地名称。
返回内容为空告知用户暂无相关结果,建议调整查询关键词

注意事项

  1. 响应时间约 1-2 分钟,调用前必须告知用户耐心等待
  2. query 越具体推荐越精准,引导用户提供:出发城市、时间、人数、预算、旅行风格
  3. Token 为极高敏感凭证,禁止在对话中打印 Token 明文
  4. 不适用于:出国签证相关问题、海外目的地咨询

Comments

Loading comments...