api

An MCP server that through www.gdacs.org provides access to web‐based disaster information systems.

我該用這個嗎

品質與安全性

A
說明品質
95%
結構描述完整度
78%
命名品質
100%
汙染風險
100%
權限相符程度
100%
協定合規性
100%

根據工具定義與協定合規性的自動化分析。

上下文成本

~908Token(工具定義)
~1.0 KB典型回應大小
中等的注意力影響(128k 上下文的 0.71%)

這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高,可用於其他工作的注意力就越少。

安裝

一鍵安裝

將以下內容加入你的 `claude_desktop_config.json` 檔案:

{
  "mcpServers": {
    "api": {
      "url": "https://api.gdacs.org/mcp"
    }
  }
}

遠端端點

https://api.gdacs.org/mcpstreamable-http

它能做什麼

工具清單

工具(8)

🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
🟢get_alert_level_info

Get explanation of GDACS alert level classifications and their meanings.

輸入結構描述

{
  "type": "object",
  "properties": {}
}
🟢search_events(eventTypes, alertLevel, country, eventname, fromDate, ...)

Search for disaster events with various filters. Returns a list of events matching the criteria, ordered by date descending.

輸入結構描述

{
  "type": "object",
  "properties": {
    "eventTypes": {
      "description": "Comma-separated event types to include (e.g., 'EQ,TC,FL'). Leave empty for all types.",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "alertLevel": {
      "description": "Alert level filter: 'green', 'orange', 'red', or combination like 'orange;red'",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "country": {
      "description": "Filter by country name or ISO code",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "eventname": {
      "description": "Filter by eventame, tipically for the TC, cyclone name",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "fromDate": {
      "description": "Start date for search range (ISO 8601 format)",
      "type": [
        "string",
        "null"
      ],
      "format": "date-time",
      "default": null
    },
    "toDate": {
      "description": "End date for search range (ISO 8601 format)",
      "type": [
        "string",
        "null"
      ],
      "format": "date-time",
      "default": null
    },
    "minSeverity": {
      "description": "Minimum severity value to filter",
      "type": [
        "number",
        "null"
      ],
      "default": null
    },
    "limit": {
      "description": "Maximum number of results to return (1-100, default 20)",
      "type": "integer",
      "default": 20
    }
  }
}
🟢get_disaster_statistics(eventTypes, alertLevel, fromDate, toDate, country)

Get disaster statistics aggregated by event type and alert level for a specified time period.

輸入結構描述

{
  "type": "object",
  "properties": {
    "eventTypes": {
      "description": "Comma-separated event types to include",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "alertLevel": {
      "description": "Alert level filter",
      "type": [
        "string",
        "null"
      ],
      "default": null
    },
    "fromDate": {
      "description": "Start date for statistics period",
      "type": [
        "string",
        "null"
      ],
      "format": "date-time",
      "default": null
    },
    "toDate": {
      "description": "End date for statistics period",
      "type": [
        "string",
        "null"
      ],
      "format": "date-time",
      "default": null
    },
    "country": {
      "description": "Country filter",
      "type": [
        "string",
        "null"
      ],
      "default": null
    }
  }
}
🟢get_events_by_country(country, daysBack, limit)

Get all disaster events affecting a specific country.

輸入結構描述

{
  "type": "object",
  "properties": {
    "country": {
      "description": "Country name or ISO3 code",
      "type": "string"
    },
    "daysBack": {
      "description": "Number of days to look back (default 30, max 365)",
      "type": "integer",
      "default": 30
    },
    "limit": {
      "description": "Maximum number of results (default 50, max 100)",
      "type": "integer",
      "default": 50
    }
  },
  "required": [
    "country"
  ]
}
🟢get_event_data(eventType, eventId)

Get detailed information about a specific disaster event. Returns event data including location, severity, affected population, and timeline.

輸入結構描述

{
  "type": "object",
  "properties": {
    "eventType": {
      "description": "Event type code: EQ (Earthquake), TC (Tropical Cyclone), FL (Flood), VO (Volcano), WF (Wildfire), DR (Drought)",
      "type": "string"
    },
    "eventId": {
      "description": "Numeric event identifier from GDACS",
      "type": "integer"
    }
  },
  "required": [
    "eventType",
    "eventId"
  ]
}
🟢get_recent_alerts(hours, minAlertLevel)

Get list of recent disaster alerts from the last N hours.

輸入結構描述

{
  "type": "object",
  "properties": {
    "hours": {
      "description": "Number of hours to look back (default 24, max 168)",
      "type": "integer",
      "default": 24
    },
    "minAlertLevel": {
      "description": "Minimum alert level: 'green', 'orange', or 'red'",
      "type": [
        "string",
        "null"
      ],
      "default": null
    }
  }
}
🟢get_active_disasters(eventTypes)

Get currently active disasters. Returns events with their geographic footprints grouped by disaster type.

輸入結構描述

{
  "type": "object",
  "properties": {
    "eventTypes": {
      "description": "Optional: semicolon-separated event types to filter (e.g., 'EQ;TC;FL')",
      "type": [
        "string",
        "null"
      ],
      "default": null
    }
  }
}
🟢get_event_types

Get list of all supported disaster event types with their descriptions.

輸入結構描述

{
  "type": "object",
  "properties": {}
}

建議的提示詞

search_research
Search for information about [topic] using api
預期的工具: search_events
find_specific
Find [specific item] using api
預期的工具: search_events
retrieve_data
Get details about [item] from api
預期的工具: get_alert_level_info
fetch_info
Fetch [information type] using api
預期的工具: get_alert_level_info
research_workflow
Search for [topic], then get detailed information about the top results using api
預期的工具: search_eventsget_alert_level_info

社群

為此伺服器評分

證據

近期觀測

已驗證未記錄版本8 個工具
已驗證未記錄版本8 個工具