# 快速開始

{% stepper %}
{% step %}

### 在開始之前，請確保你具備以下條件：

<table data-view="cards"><thead><tr><th></th><th data-type="image"></th><th></th></tr></thead><tbody><tr><td><a href="https://dashboard.talordata.com/reg"><strong>Talordata账户</strong></a></td><td></td><td><em>注册帳戶即可獲取1000次免費SERP API請求。</em></td></tr><tr><td><a href="https://dashboard.talordata.com/scraping/serp-api/api-token"><strong>Talordata API Token</strong></a><a href="https://dashboard.talordata.com/scraping/serp-api/api-token"> </a></td><td></td><td><em>在SERP API儀錶板獲取API密钥。</em></td></tr></tbody></table>

{% endstep %}

{% step %}

### 使用以下端點連接到遠程MCP服務器

```
{
  "mcpServers": {
    "talordata": {
      "type": "http",
      "url": "https://mcp.talordata.net/YOUR_API_TOKEN/mcp"
    }
  }
}
```

{% endstep %}
{% endstepper %}

### 認證

支持三種方法：

1. **路徑型（推薦）:** <kbd><mark style="color:red;">/{YOUR\_API\_TOKEN}/mcp<mark style="color:red;"></kbd>
2. **Authorization 头:** <kbd><mark style="color:red;">Authorization: Bearer YOUR\_API\_TOKEN<mark style="color:red;"></kbd>&#x20;
3. **自定義頭:** <kbd><mark style="color:red;">X-Talor-Serp-Token: YOUR\_API\_TOKEN<mark style="color:red;"></kbd>

> 示例:

#### 路徑型

```
curl "https://mcp.talordata.net/your_token/mcp" -d '...'
```

#### Authorization 头

```
curl "https://mcp.talordata.net/mcp" -H "Authorization: Bearer your_token" -d '...'
```

#### 自定義頭

```
curl "https://mcp.talordata.net/mcp" -H "X-Talor-Serp-Token: your_token" -d '...'
```

{% hint style="info" %}
注意: GET / 和 GET /healthz 端点无需认证。
{% endhint %}

### 工具

#### **1.list\_engines**&#x20;

列出所有支持的搜索引擎、分類和資源URI。

參數：無

示例: <kbd>{"name": "list\_engines"}</kbd>

#### **2.search**&#x20;

執行 SERP 搜索请求。使用 <kbd>talor://engines</kbd> 下的資源查看各引擎的具體參數後再調用此工具。

参数:&#x20;

<kbd>engine</kbd> (可選): 引擎标识，例如 google, google\_images, bing\_images, duckduckgo。&#x20;

如果不提供，使用 params 中的 engine 字段或默認引擎。&#x20;

<kbd>params</kbd> (必填): 引擎特定參數對象，需匹配對應引擎的Schema資源定義。&#x20;

<kbd>json</kbd>（可選）：上游响应格式。 json->1，json-html->2，html->3。

<kbd>response\_mode</kbd> (可選): complete 返回完整响应，compact 移除常見元数据字段。默認：complete

**支持的搜索引擎（共 33 個）:**&#x20;

<details>

<summary>Google (25)</summary>

Search, Finance, Finance Markets, Flights, Hotels, Images, Jobs, Lens, Local, Maps, News, Patents, Patents Details, Play Search, Play Books, Play Games, Play Movies, Play Product, Scholar, Scholar Author, Scholar Cite, Shopping, Trends, Videos, Web&#x20;

</details>

<details>

<summary>Bing (6)</summary>

Search, Images, Maps, News, Shopping, Videos&#x20;

</details>

<details>

<summary>Yandex (1)</summary>

Search&#x20;

</details>

<details>

<summary>DuckDuckGo (1)</summary>

Search

</details>

> 示例:&#x20;

```
{"name": "search", "arguments": {"engine": "google", "params": {"q": "coffee shops", "location": "Austin, TX"}}} 
{"name": "search", "arguments": {"engine": "google", "params": {"q": "weather in London"}}} 
{"name": "search", "arguments": {"engine": "google", "params": {"q": "AAPL stock"}}} 
{"name": "search", "arguments": {"engine": "duckduckgo", "params": {"q": "car"}}} 
{"name": "search", "arguments": {"engine": "bing", "params": {"q": "latest news"}, "response_mode": "compact"}}
```

結果類型：答案框、自然結果、新聞、圖片、購物、知識卡片——由上游API自動檢測並格式化。

#### **3.history**&#x20;

査詢SERP使用歷史記錄。

参数:&#x20;

<kbd>page</kbd> (可選): 頁碼，默認1

<kbd>page\_size</kbd> (可選): 每頁大小，常用20/50/100

<kbd>search\_query</kbd>(可選): 按搜索关键词篩選&#x20;

<kbd>search\_engine</kbd> (可選): 按搜索引擎顯示名稱篩選

<kbd>status</kbd> (可選): 狀態篩選。 可選值：all, success, error&#x20;

<kbd>start\_time</kbd> (可選): 起始時間（Unix時間戳，秒）

<kbd>end\_time</kbd> (可選): 結束時間（Unix時間戳，秒）

<kbd>timezone</kbd> (可選): 時區，例如 Asia/Shanghai 或 +08:00

> 示例: &#x20;

```
{"name": "history", "arguments": {"page": 1, "page_size": 20}} 
{"name": "history", "arguments": {"search_query": "coffee", "status": "success"}} 
{"name": "history", "arguments": {"start_time": 1700000000, "end_time": 1700100000, "timezone": "Asia/Shanghai"}}
```

#### **4.statistics**&#x20;

査詢SERP使用統計数据。

參數：

<kbd>start\_date</kbd> (必填): 起始日期，格式 YYYY-MM-DD&#x20;

<kbd>end\_date</kbd> (必填): 結束日期，格式 YYYY-MM-DD&#x20;

<kbd>engines</kbd> (可選):  引擎篩選，逗號分隔的字符串或字符串數組

<kbd>timezone</kbd> (可選): 時區偏移，例如 +00:00, +08:00, -05:00

> 示例:&#x20;

```
{"name": "statistics", "arguments": {"start_date": "2025-01-01", "end_date": "2025-01-31"}} 
{"name": "statistics", "arguments": {"start_date": "2025-01-01", "end_date": "2025-01-31", "engines": "google,bing"}} 
{"name": "statistics", "arguments": {"start_date": "2025-01-01", "end_date": "2025-01-31", "timezone": "+08:00"}}
```

### MCP 資源

* 索引資源： <kbd>talor://engines</kbd>&#x20;

&#x20;     返回所有引擎的列表、分類和資源URI。

* 引擎Schema資源： <kbd>talor://engines/{engine\_key}</kbd>&#x20;

&#x20;     返回特定引擎的完整參數Schema定義，用於在調用search工具前瞭解可用參數。

### HTTP 路由

1. <kbd>GET /</kbd> 返回服務基本信息（無需認證）&#x20;
2. <kbd>GET /healthz</kbd> 健康檢查（無需認證）
3. <kbd>POST /mcp</kbd> MCP 協议端點（需要認證）

如需更多協助，請透過 **線上客服** 或電郵 **<support@talordata.com>** 聯絡我們。


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.talordata.com/cn-tw/serp-api/mcp-server/quick-start.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.