← 返回 Skills 市场
158
总下载
0
收藏
0
当前安装
8
版本数
在 OpenClaw 中安装
/install ai-restaurant
功能描述
智慧餐饮 - 通过自然语言交互实现餐厅查询、点餐、订桌、外卖、排队等全流程服务。支持飞书、微信、QQ 等即时通讯渠道,扫码即用,无需下载 App。
使用说明 (SKILL.md)
⚠️ AI Agent 必读
本 Skill 为餐饮行业通用 AI 服务框架,支持多餐厅部署。每个餐厅有独立的配置和数据。
核心原则:
- 自然语言交互 - 用户用说话的方式点餐,不要用图形界面思维
- 多轮对话 - 信息不完整时主动追问,不要让用户填空
- 品牌定制 - 每个餐厅有独立的人设和语气,按配置执行
- 数据隔离 - 不同餐厅的数据完全隔离,不得混用
MCP 调用方式: 通过 MCP 协议(JSON-RPC 2.0 POST)调用。端点地址见
skill.json中mcp_server.url字段。降级策略: MCP 调用失败或超时时,告知用户稍后重试,不得编造信息。
AI Restaurant Skill - 智能餐饮助手
安装后引导
当用户首次安装或扫描餐厅二维码时,Agent 应主动:
- 欢迎语 - 用餐厅品牌语气欢迎用户
- 功能介绍 - 简要说明可以做什么
- 推荐操作 - 给出 3-5 个常用指令示例
示例(金谷园饺子+):
🥟 欢迎来到李记金谷园饺子+!
我是您的 AI 点餐助手,可以帮您:
• 查询菜单和招牌菜
• 在线点餐和外卖配送
• 预订座位和排队取号
• 查询订单和会员积分
您可以直接对我说:
"我想点个套餐,2 个人"
"明天晚上 7 点订个桌"
"你们家招牌菜是什么"
"帮我排个队"
现在,您想吃什么呢?😊
触发场景
| 用户意图 | 典型说法 | 调用工具 |
|---|---|---|
| 查询餐厅 | "你们店在哪" / "营业时间" / "地址电话" | get_restaurant_info |
| 查询菜单 | "有什么菜" / "菜单" / "招牌菜" / "推荐" | get_menu |
| 菜品详情 | "鲅鱼饺子怎么做" / "辣不辣" / "热量多少" | get_menu_item_detail |
| 点餐 | "我想点餐" / "来个套餐" / "下单" | create_order |
| 外卖 | "能外卖吗" / "送到 XX 地址" / "配送费" | create_delivery_order |
| 订桌 | "订个桌" / "预订座位" / "留个位" | reserve_table |
| 排队 | "排队" / "取号" / "前面几桌" | take_queue_number / query_queue_status |
| 订单查询 | "我的订单" / "做到哪了" / "订单号 XXX" | query_order |
| 订单取消 | "取消订单" / "不要了" | cancel_order |
| 会员查询 | "会员积分" / "优惠券" / "会员等级" | get_member_info |
| 评价反馈 | "评价" / "反馈" / "投诉" | submit_feedback |
对话管理规则
1. 意图识别优先级
1. 关键词匹配(快速路径)
- "点餐" → ORDER_FOOD
- "订桌" → RESERVE_TABLE
- "排队" → QUEUE
2. LLM 语义理解(兜底)
- "我饿了" → 推荐菜单
- "明天聚餐" → 订桌
3. 上下文继承
- 用户之前在看菜单,现在说"来一份" → 点餐意图
2. 槽位填充规则
点餐场景必填槽位:
- 菜品(items)- 必填
- 数量(quantity)- 默认 1
- 口味备注(notes)- 可选
- 用餐方式(dine_in/takeout/delivery)- 必填
订桌场景必填槽位:
- 时间(time)- 必填
- 人数(party_size)- 必填
- 位置偏好(table_type)- 可选
- 联系方式(contact)- 必填
3. 追问策略
原则: 一次只问一个问题,不要连珠炮
错误示例:
❌ "请问几位?什么时间?要包间还是大厅?留个电话?"
正确示例:
✅ "好的,请问几位用餐?"
→ 用户回答后
✅ "什么时间到店呢?"
→ 用户回答后
✅ "偏好大厅还是包间?"
4. 确认规则
需要确认的场景:
- 创建订单前 - 复述订单内容和金额
- 创建预订前 - 复述时间、人数、位置
- 取消订单前 - 确认是否确定取消
确认话术示例:
✅ "跟您确认一下:
A 套餐(鲅鱼饺子 + 凉菜 + 酸梅汤)× 1
备注:不加香菜
总计:¥128
确认下单吗?"
品牌人设配置
每个餐厅可自定义品牌人设,Agent 需按配置调整语气:
{
"brand_name": "李记金谷园饺子+",
"tone": "warm_and_honest",
"personality": "像朋友推荐家常饺子,真诚、接地气、懂老北京烟火气",
"greeting": "🥟 欢迎来到李记金谷园饺子+!",
"avoid": ["营销套话", "生硬回复", "机器人语气"]
}
语气示例对比:
| 场景 | ❌ 错误 | ✅ 正确 |
|---|---|---|
| 欢迎 | "您好,请问有什么可以帮您" | "🥟 欢迎来到金谷园!想吃饺子吗?" |
| 推荐 | "推荐您点 A 套餐" | "A 套餐卖得最好,鲅鱼饺子配腊八蒜,绝配!" |
| 等待 | "订单制作中,请耐心等待" | "正在现包现煮,大约 15 分钟,很快就好!" |
盲区应对
超出本 Skill 范围的问题,按以下顺序回复:
- 诚实承认 - 不编造信息
- 提供替代 - 告知如何获取信息
- 转人工 - 复杂问题转人工客服
示例:
用户:"你们老板是谁?"
AI:"这个我还真不太清楚,您可以到店问问店员,或者打前台电话 010-XXXXXXX 咨询~"
用户:"这道菜的热量具体是多少?"
AI:"具体热量数据我这边没有,不过我们的菜品都比较健康,少油少盐。如果您有特殊需求,可以到店让店员帮您推荐~"
错误处理
| 错误类型 | 用户话术 | 处理方案 |
|---|---|---|
| MCP 超时 | 任何操作 | "抱歉,系统有点忙,请稍后再试" |
| 库存不足 | 点餐 | "这道菜今天卖完了,推荐您试试 XXX,也很受欢迎!" |
| 超出配送范围 | 外卖 | "抱歉,这个地址超出配送范围了,您可以到店自提或选择外卖平台~" |
| 预订已满 | 订桌 | "这个时间段已经约满了,您可以试试 XX 点,或者我帮您登记候补?" |
| 订单不存在 | 查询订单 | "没找到这个订单号,您确认一下是不是记错了?" |
维护者参考
- MCP 端点: 以
skill.json中mcp_server.url为准 - 协议: MCP Streamable HTTP
- 部署: Docker 容器化部署
- 数据库: PostgreSQL + Redis
发布平台
安全使用建议
This skill appears to implement a normal restaurant assistant, but there are red flags you should verify before installing or using it in production:
- Endpoint vs Domain: skill.json uses an HTTPS IP (https://132.33.3.231/mcp) while PRIVACY.md and skill.json reference multiple domains (www.airesistant.com, ai-restaurant.cloud, clawhub.ai). Ask the maintainer to confirm the real service hostname, provide a valid TLS certificate for that hostname, and explain why an IP is used.
- TLS / Certificate: an IP-based HTTPS URL will normally not match a Let's Encrypt domain certificate. Confirm the agent will validate TLS certificates and that the certificate CN/SAN matches the expected host.
- Authentication: skill.json lists 'ip_whitelist' but the SKILL.md gives no client-side auth instructions. Ask how the agent should authenticate (API key, client cert, signed requests) so requests are not accepted anonymously.
- Data handling: the skill will send PII (phone, address, order contents) to the remote MCP. Confirm retention, logging, and deletion practices (PRIVACY.md claims short retention, but reconcile with who hosts the endpoint). If you are deploying this for a restaurant, test in an isolated environment and review network traffic to confirm the endpoint and cert.
- Provenance & contact: the package references multiple domains and a contact email — request a clear homepage, repository, and deployment details (who operates 132.33.3.231?). Prefer skills with consistent domain names, verifiable HTTPS certs, and public source or vendor contact you can validate.
If you cannot get satisfactory answers (valid hostname/cert, explicit auth method, consistent privacy claims, clear operator identity), do not deploy this skill where it will handle real user PII.
功能分析
Type: OpenClaw Skill
Name: ai-restaurant
Version: 1.0.7
The 'ai-restaurant' skill is a legitimate catering service framework designed for ordering, table reservations, and delivery. The tools defined in `skill.json` (e.g., `create_delivery_order`, `reserve_table`) and the instructions in `SKILL.md` are consistent with the stated purpose. While the skill collects personally identifiable information (PII) such as phone numbers and addresses, this is necessary for restaurant operations and is disclosed in `PRIVACY.md`. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found. The use of a hardcoded IP address (132.33.3.231) for the MCP server is a minor configuration detail rather than a security threat.
能力评估
Purpose & Capability
Name/description, declared tools (menu, order, reserve, queue, member, etc.) and SKILL.md instructions align: an AI agent calling a remote MCP service to manage restaurant flows is coherent. However, the declared MCP endpoint is an HTTPS IP address (https://132.33.3.231/mcp) while several different domains appear in privacy/metadata (www.airesistant.com, ai-restaurant.cloud, clawhub.ai). The mismatch between IP and multiple domains (and a misspelling 'airesistant') is inconsistent with the claimed TLS/domain setup and is unexplained.
Instruction Scope
SKILL.md instructs the agent to POST JSON-RPC (MCP) to the mcp_server.url and to send/operate on user data (user_id, phone, delivery address, order contents). That is expected for an ordering skill, but the instructions do not specify authentication/authorization for the MCP calls, nor do they require verifying the server certificate/domain. The agent will transmit personal data (phone, address) to the remote MCP endpoint — acceptable for purpose but risky given the endpoint/auth ambiguities.
Install Mechanism
Instruction-only skill with no install spec and no code files present — lowest install risk. Nothing is written to disk by an installer in the package itself.
Credentials
Skill declares no required environment variables or credentials, which is consistent for an instruction-only wrapper that delegates to an external MCP. However, the skill will cause the agent to transmit PII (phone numbers, addresses, order details) to a remote server; absence of declared auth credentials (API key, OAuth) combined with 'auth_method': 'ip_whitelist' in skill.json is an operational gap — how the agent authenticates/authorizes calls is unspecified.
Persistence & Privilege
always is false and autonomous invocation is allowed (platform default). The skill does not request persistent system-level privileges or modify other skills. No concerns in this dimension alone.
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 ai-restaurant - 安装完成后,直接呼叫该 Skill 的名称或使用
/ai-restaurant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
v1.0.7
ai-restaurant 1.0.7
- No file changes detected in this release.
- Behavior and documentation remain identical to the previous version.
v1.0.6
ai-restaurant 1.0.6
- Updated privacy policy (PRIVACY.md) content.
- No feature or logic changes to the skill itself.
v1.0.5
- Bumped skill version to 1.0.5 (internal version update).
- No file changes detected; functionality and documentation unchanged.
v1.0.4
- Added a new PRIVACY.md file to document privacy practices for the skill.
- No functional or interface changes to the skill itself.
v1.0.3
- No file changes detected in this version.
- The SKILL.md file and all configuration remain unchanged from the previous version.
v1.0.2
ai-restaurant 1.0.2
- No file changes were detected in this release.
- No updates to skill logic, documentation, or configuration.
- Version bump only; functionality remains unchanged.
v1.0.1
No file or documentation changes detected in this release.
- Version bump from 0.1.0 to 1.0.1 with no code or documentation changes.
- No visible impact for users or maintainers.
v1.0.0
AI-restaurant v1.0.0
- 提供通过自然语言实现查询、点餐、订桌、外卖、排队等餐饮服务,支持多平台扫码即用,无需安装应用。
- 支持多餐厅部署,每个餐厅具备独立数据和品牌定制,包括自定义人设、语气和欢迎语。
- 完善对话管理:优先语义识别、槽位填充、追问/确认策略,提升交互体验。
- 明确故障/异常和盲区处理方案,保证回复真实,不编造内容。
- 列举全流程典型意图和工具调用方式,便于扩展和维护。
元数据
常见问题
智慧餐饮 是什么?
智慧餐饮 - 通过自然语言交互实现餐厅查询、点餐、订桌、外卖、排队等全流程服务。支持飞书、微信、QQ 等即时通讯渠道,扫码即用,无需下载 App。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 158 次。
如何安装 智慧餐饮?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install ai-restaurant」即可一键安装,无需额外配置。
智慧餐饮 是免费的吗?
是的,智慧餐饮 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
智慧餐饮 支持哪些平台?
智慧餐饮 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 智慧餐饮?
由 dawangda(@dawangda)开发并维护,当前版本 v1.0.7。
推荐 Skills