Google Research Tools

Google Research Tools: google Search API

Google Research Tools: google Search API.

MCP tool name

nexscope_data_ai_search_google_search

MCP endpoint

https://api.nexscope.ai/mcp

JSON-RPC method

tools/call

Authentication

Use OAuth 2.1 Authorization Code with PKCE and the mcp:tools scope. API keys remain supported as a compatibility option.

Authorization: Bearer nk-xxxxxxxxxxxxxxxxx

If authentication fails, ask the user to open API Access and copy a valid key.

Call this MCP tool

Initialize an MCP 2025-11-25 Streamable HTTP session at https://api.nexscope.ai/mcp, then use tools/call.

{
  "jsonrpc": "2.0",
  "id": "data-ai-search-google-search-1",
  "method": "tools/call",
  "params": {
    "name": "nexscope_data_ai_search_google_search",
    "arguments": {
      "keyword": "best wireless earbuds 2026"
    }
  }
}

MCP 2025-11-25 · Streamable HTTP

Progress is delivered on the original POST SSE response. GET SSE carries notifications; reconnecting must query existing asynchronous work rather than submit again.

Arguments

Name	Type	Required	Description
`keyword`	`string`	required	Google search keywords as `q=` parameter to initiate Google AI Mode search. Only a single round of dialogue is supported; if you need to ask questions, the agent needs to summarize the key information of the previous round of AI overview and splice new questions, and then initiate a request as a new `keyword`. Example: {}

Response fields

Treat structuredContent as the programmatic API response payload.

Name	Type	Required	Description
`stdout`	`string`	optional	AI overview text, in Markdown format; output the AI overview points and reference source links for each question paragraph by paragraph in order of questions. Example: {}
`sourceUrl`	`string`	optional	Crawl the target URL and eventually send it to Google's search page address. Example: {}
`resultsNum`	`integer`	optional	The number of AI overview blocks; greater than 0 means that the page renders AI Overview, and 0 means that the keyword does not trigger AI Overview. Example: {}
`code`	`string`	optional	Business status code, successful is `"200"`, corresponding to the numerical version of `errcode`. Example: {}
`errcode`	`integer`	optional	Business status code; `200` indicates success, other values indicate business errors. Example: {}
`costTime`	`integer`	optional	The interface takes time in milliseconds. Example: {}
`costToken`	`integer`	optional	This call consumes the number of Tokens; it will be billed only if the upstream returns successfully. Example: {}
`taskId`	`string`	optional	The identification of this crawling task returned by the upstream. Example: {}
`type`	`string`	optional	The rendering style, fixed to `stdoutWorkbenches`, is rendered in Markdown format with the `stdout` field. Example: {}

Examples

Example arguments

{
  "keyword": "best wireless earbuds 2026"
}

Example structuredContent

{
  "code": 0,
  "cost": "-1",
  "data": {},
  "msg": null,
  "time": "2026-01-01 00:00:00",
  "traceId": null,
  "ts": "0"
}