DealScore

For AI agents: grade a used car's asking price vs live market comps.

我該用這個嗎

品質與安全性

B
說明品質
80%
結構描述完整度
63%
命名品質
80%
汙染風險
100%
權限相符程度
100%
協定合規性
100%

發現項目(2)

  • LOWTool 'score_deal' description lacks action verb在 score_deal 中
  • LOWTool 'explain_grade' description lacks action verb在 explain_grade 中

根據工具定義與協定合規性的自動化分析。

上下文成本

~347Token(工具定義)
~656 B典型回應大小
極小的注意力影響(128k 上下文的 0.27%)

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": {
    "dealscore": {
      "url": "https://cardealscore.com/api/mcp"
    }
  }
}

遠端端點

https://cardealscore.com/api/mcpstreamable-http

它能做什麼

工具清單

工具(3)

🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
🟡score_deal(vin, year, make, model, trim, ...)

Grade a car deal: compare asking price to expected market value. ZIP is optional. Same inputs/outputs as POST /api/score.

輸入結構描述

{
  "type": "object",
  "properties": {
    "vin": {
      "type": "string"
    },
    "year": {
      "type": "integer",
      "minimum": 1980,
      "maximum": 2027
    },
    "make": {
      "type": "string"
    },
    "model": {
      "type": "string"
    },
    "trim": {
      "type": "string"
    },
    "asking_price": {
      "type": "number",
      "exclusiveMinimum": 0
    },
    "mileage": {
      "type": "number",
      "minimum": 0
    },
    "zip": {
      "type": "string",
      "pattern": "^\\d{5}$"
    }
  },
  "required": [
    "asking_price",
    "mileage"
  ],
  "$schema": "https://json-schema.org/draft/2020-12/schema"
}
🟢explain_grade(delta_pct)

Explain the score_v1 rubric grade for a delta_pct. Same as POST /api/explain.

輸入結構描述

{
  "type": "object",
  "properties": {
    "delta_pct": {
      "type": "number"
    }
  },
  "required": [
    "delta_pct"
  ],
  "$schema": "https://json-schema.org/draft/2020-12/schema"
}
🟢health

Service health, coverage, cache policy, and limitations. Same as GET /api/health.

輸入結構描述

{
  "type": "object",
  "properties": {},
  "$schema": "https://json-schema.org/draft/2020-12/schema"
}

社群

為此伺服器評分

證據

近期觀測

已驗證未記錄版本3 個工具