Safe Bite

Restaurant inspections for Chicago, NYC, Minneapolis, Milwaukee, Austin, and Denver.

我该使用它吗

质量与安全性

A
描述质量
95%
模式完整度
95%
命名质量
100%
投毒风险
100%
权限匹配度
100%
协议合规性
100%

基于对工具定义和协议合规性的自动分析。

上下文开销

~946token 数(工具定义)
~2.0 KB典型响应大小
对注意力有中等影响(占 128k 上下文窗口的 0.74%)

这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高,可用于其他任务的注意力就越少。

安装

一键安装

将以下内容添加到你的 `claude_desktop_config.json` 文件中:

{
  "mcpServers": {
    "safe-bite": {
      "url": "https://www.safebiter.com/mcp"
    }
  }
}

远程端点

https://www.safebiter.com/mcpstreamable-http

它能做什么

工具清单

工具(4)

🟢 只读🟡 写入🔴 删除⚪ 未知
🟢get_milwaukee_inspections(limit, offset)

Get Milwaukee restaurant food inspection data with full pagination support. Response includes pagination metadata (total_count, offset, limit, has_more, next_offset) to guide further requests. To page through all records, repeat the call passing next_offset until has_more is false.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "default": 10,
      "description": "Number of records per request (1-500). Use 500 to get maximum data per call.",
      "maximum": 500,
      "minimum": 1,
      "type": "integer"
    },
    "offset": {
      "default": 0,
      "description": "Number of records to skip, for pagination. Pass the previous response's next_offset to get the next page. Default 0.",
      "minimum": 0,
      "type": "integer"
    }
  }
}
🟢get_city_inspections(city, limit, offset)

Get restaurant inspection records for one city: chicago, nyc, minneapolis, milwaukee, austin, or denver. Page by repeating the call with the previous response's next_offset until has_more is false. The response includes total_count, offset, limit, has_more, and next_offset.

输入模式

{
  "type": "object",
  "properties": {
    "city": {
      "default": "chicago",
      "description": "City to query: chicago, nyc, minneapolis, milwaukee, austin, or denver.",
      "enum": [
        "chicago",
        "nyc",
        "minneapolis",
        "milwaukee",
        "austin",
        "denver"
      ],
      "type": "string"
    },
    "limit": {
      "default": 10,
      "description": "Number of records per request (1-500). Use 500 to get maximum data per call for large datasets.",
      "maximum": 500,
      "minimum": 1,
      "type": "integer"
    },
    "offset": {
      "default": 0,
      "description": "Number of records to skip, for pagination. Pass the previous response's next_offset to get the next page. Default 0.",
      "minimum": 0,
      "type": "integer"
    }
  },
  "required": [
    "city"
  ]
}
🟢search_restaurants(city, history_n, limit, query, threshold)

Fuzzy-search restaurants by name across Chicago, NYC, Minneapolis, Milwaukee, Austin, and Denver. Tolerates misspellings and partial names. Returns one entry per restaurant with its recent inspections, newest first, and a similarity score.

输入模式

{
  "type": "object",
  "properties": {
    "city": {
      "default": "all",
      "description": "Optional city scope.",
      "enum": [
        "all",
        "chicago",
        "nyc",
        "minneapolis",
        "milwaukee",
        "austin",
        "denver"
      ],
      "type": "string"
    },
    "history_n": {
      "default": 3,
      "description": "How many recent inspections to include per restaurant. Default 3.",
      "maximum": 20,
      "minimum": 1,
      "type": "integer"
    },
    "limit": {
      "default": 10,
      "description": "Max number of matches to return (1-50).",
      "maximum": 50,
      "minimum": 1,
      "type": "integer"
    },
    "query": {
      "description": "Restaurant name or partial name (e.g., 'giordanos', 'shake shak').",
      "type": "string"
    },
    "threshold": {
      "default": 0.6,
      "description": "Fuzzy match threshold between 0 and 1 (higher = stricter).",
      "maximum": 1,
      "minimum": 0,
      "type": "number"
    }
  },
  "required": [
    "query"
  ]
}
🟢find_nearby_restaurants(address, city, history_n, latitude, limit, ...)

Find restaurants within a given radius of a location and return each restaurant's last 3 inspections (newest first). Accepts either explicit latitude/longitude or an address string (geocoded via Google Maps). Default radius is a quarter mile (0.25).

输入模式

{
  "type": "object",
  "properties": {
    "address": {
      "description": "Optional street address; resolved to lat/lng via Google Maps Geocoding API. Used only if latitude/longitude are not supplied.",
      "type": "string"
    },
    "city": {
      "default": "all",
      "description": "Optional city filter.",
      "enum": [
        "all",
        "chicago",
        "nyc",
        "minneapolis",
        "milwaukee",
        "austin",
        "denver"
      ],
      "type": "string"
    },
    "history_n": {
      "default": 3,
      "description": "How many recent inspections to include per restaurant. Default 3.",
      "maximum": 20,
      "minimum": 1,
      "type": "integer"
    },
    "latitude": {
      "description": "Center latitude in decimal degrees. Required if no address is provided.",
      "type": "number"
    },
    "limit": {
      "default": 20,
      "description": "Max restaurants to return (1-100).",
      "maximum": 100,
      "minimum": 1,
      "type": "integer"
    },
    "longitude": {
      "description": "Center longitude in decimal degrees. Required if no address is provided.",
      "type": "number"
    },
    "radius_miles": {
      "default": 0.25,
      "description": "Search radius in miles. Defaults to 0.25 (a quarter mile).",
      "maximum": 25,
      "minimum": 0.01,
      "type": "number"
    }
  }
}

社区

评价此服务器

证据

最近观测

已验证未记录版本4 个工具