Talordata Mcp Server

@Talordata

About Talordata Mcp Server

Talordata MCP Server is a remote Model Context Protocol server for Talordata SERP API. It connects multi-engine SERP capabilities with MCP-compatible clients, giving your AI real-time web search, request history lookup, and usage analytics.

#mcp-server #search-data #ai-agent

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "talordata": {
      "url": "https://your-domain.com:8800/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_USER_TOKEN"
      }
    }
  }
}

Tools

`tools/history.go`

Overview

What is Talordata Mcp Server?

Talordata Mcp Server is an MCP server that gives MCP clients access to Talor SERP search, history, and statistics. It is built with mark3labs/mcp-go, backed by local engine schemas, and designed for streamable HTTP deployment.

How to use Talordata Mcp Server?

Clone the repository, run go mod tidy && go run ., and configure via configs/config.yaml. The server exposes HTTP endpoints at /mcp and /{user-token}/mcp; clients connect by providing a URL and an authorization header containing their Talor SERP user token.

Key features of Talordata Mcp Server

Dynamically loads all engine schemas from local engines/*.json
Exposes engine index and raw engine schemas through MCP resources
Proxies Talor SERP search requests through the search tool
Proxies usage history through the history tool
Proxies usage statistics through the statistics tool
Supports per-request user token forwarding instead of a fixed upstream token

Use cases of Talordata Mcp Server

Perform live SERP searches across multiple engines (e.g., google_web, google_images)
Query usage history with filters such as date range, engine, and status
Retrieve usage statistics within a specified date period
Discover available engines and their parameter schemas dynamically

FAQ from Talordata Mcp Server

How does authentication work?

Each user provides their own Talor SERP token per request, delivered via Authorization: Bearer <user-token> header, X-Talor-Serp-Token header, or in the URL path as /{user-token}/mcp. Query string token passing is not supported.

What transport does the server use?

The server is designed for cloud deployment with streamable-http. It listens on a configurable address and exposes GET /healthz and POST / GET / DELETE /mcp endpoints.

Where does the server load engine schemas from?

Engine schemas are loaded from local engines/*.json files at startup. The engine index is served under talor://engines and individual schemas under talor://engines/<engine>.

What dependencies or runtime are required?

The server is written in Go. Running it locally requires Go installed, and you execute go mod tidy and go run . to start. A sample configuration file (configs/config.yaml) is provided.

What are the configurable timeouts?

The server allows setting timeout_ms (upstream timeout, default 150000) and shutdown_timeout_ms (graceful shutdown timeout, default 10000) in the configuration file.

Frequently asked questions

How does authentication work?

Each user provides their own Talor SERP token per request, delivered via `Authorization: Bearer <user-token>` header, `X-Talor-Serp-Token` header, or in the URL path as `/{user-token}/mcp`. Query string token passing is not supported.

What transport does the server use?

The server is designed for cloud deployment with `streamable-http`. It listens on a configurable address and exposes `GET /healthz` and `POST / GET / DELETE /mcp` endpoints.

Where does the server load engine schemas from?

Engine schemas are loaded from local `engines/*.json` files at startup. The engine index is served under `talor://engines` and individual schemas under `talor://engines/<engine>`.

What dependencies or runtime are required?

The server is written in Go. Running it locally requires Go installed, and you execute `go mod tidy` and `go run .` to start. A sample configuration file (`configs/config.yaml`) is provided.

What are the configurable timeouts?

The server allows setting `timeout_ms` (upstream timeout, default 150000) and `shutdown_timeout_ms` (graceful shutdown timeout, default 10000) in the configuration file.