US FDA Recalls (Food, Drug, Device) MCP
FDA food, drug and medical-device recalls by keyword, class and state.
我該用這個嗎
品質與安全性
A
根據工具定義與協定合規性的自動化分析。
上下文成本
~290Token(工具定義)
~700 B典型回應大小
極小的注意力影響(128k 上下文的 0.23%)
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": {
"us-food-drug-recalls-mcp": {
"url": "https://us-food-drug-recalls-mcp.donniertf.workers.dev/mcp"
}
}
}遠端端點
https://us-food-drug-recalls-mcp.donniertf.workers.dev/mcpstreamable-http它能做什麼
工具清單
工具(3)
🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
⚪latest_recalls(keyword, type)
FREE. Latest 5 FDA recalls (food, drug or device) matching a keyword (product, firm, allergen, pathogen).
輸入結構描述
{
"type": "object",
"properties": {
"keyword": {
"type": "string"
},
"type": {
"type": "string",
"description": "food (default) | drug | device"
}
}
}🟢recall_search(keyword, classification, state, since, types, ...)
PRO. Up to 100 recalls across food, drugs and devices with filters: keyword, class (I/II/III), state, since date.
輸入結構描述
{
"type": "object",
"properties": {
"keyword": {
"type": "string"
},
"classification": {
"type": "string"
},
"state": {
"type": "string"
},
"since": {
"type": "string",
"description": "YYYY-MM-DD"
},
"types": {
"type": "array",
"items": {
"type": "string"
}
},
"access_token": {
"type": "string",
"description": "Subscriber token from /success (or Authorization: Bearer)"
}
}
}🟢watchlist(items, days, access_token)
PRO. Check up to 50 suppliers/brands/products against recalls in the last N days (default 90).
輸入結構描述
{
"type": "object",
"properties": {
"items": {
"type": "array",
"items": {
"type": "string"
}
},
"days": {
"type": "number"
},
"access_token": {
"type": "string",
"description": "Subscriber token from /success (or Authorization: Bearer)"
}
},
"required": [
"items"
]
}社群
證據
近期觀測
已驗證未記錄版本3 個工具