Serper MCP server for AI agents
Let Claude, ChatGPT or your own agent read Serper: 3 typed functions through one MCP server. No Serper API key needed.
- Site
- serper.dev
- Functions
- 3
- Sign-in
- Not needed
- Works with
- MCP, HTTP, Node, Python
Connect
Just Serper
An MCP server for Serper alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.
https://api.bowmark.ai/mcp/site/serperEvery site
One MCP server for Serper and every other site Bowmark covers. Your agent looks up what it needs and calls it.
https://api.bowmark.ai/mcpSet it up in your clientBoth need a Bowmark API key. The quickstart gets you one and a first result in a few minutes.
Functions
Open a function to see its arguments, what it returns and how to call it.
Read 3
searchRuns a Google search through Serper's API and returns Google's organic results.Arguments
Name Type Required querystringYes opts.numnumber1-100, default 10.No opts.glstringCountry, e.g. "us".No opts.hlstringLanguage, e.g. "en".No Returns
SerperSearchResultinterface SerperOrganicResult { position: number; title: string; link: string; // the destination url snippet: string | null; date: string | null; } interface SerperSearchResult { query: string; results: SerperOrganicResult[]; // empty only when Google matched nothing credits: number | null; // Serper credits this search spent warnings: string[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/serper/searchCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.serper.search("site:cal.com \"Chris Field\"");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.serper.search("site:cal.com \"Chris Field\""))curl -X POST https://api.bowmark.ai/v1/sites/serper/search \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"query":"site:cal.com \"Chris Field\""}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.serper.search("site:cal.com \"Chris Field\"");searchGoogleAlias for search(), kept for backward compatibility.Arguments
Name Type Required querystringYes opts.numnumber1-100, default 10.No opts.glstringCountry, e.g. "us".No opts.hlstringLanguage, e.g. "en".No Returns
SerperSearchResultinterface SerperOrganicResult { position: number; title: string; link: string; // the destination url snippet: string | null; date: string | null; } interface SerperSearchResult { query: string; results: SerperOrganicResult[]; // empty only when Google matched nothing credits: number | null; // Serper credits this search spent warnings: string[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/serper/searchGoogleCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.serper.searchGoogle("site:github.com octocat");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.serper.searchGoogle("site:github.com octocat"))curl -X POST https://api.bowmark.ai/v1/sites/serper/searchGoogle \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"query":"site:github.com octocat"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.serper.searchGoogle("site:github.com octocat");searchShoppingSearches Google Shopping through Serper's API and returns Google Shopping's product rows.Arguments
Name Type Required querystringYes opts.numnumber1-100, default 10.No opts.glstringCountry, e.g. "us".No opts.hlstringLanguage, e.g. "en".No Returns
SerperShoppingSearchResultinterface SerperShoppingResult { position: number; title: string; source: string | null; // the seller, e.g. "Best Buy" link: string; // the product / offer page price: string | null; // as Google prints it, e.g. "$249.99" imageUrl: string | null; rating: number | null; ratingCount: number | null; productId: string | null; } interface SerperShoppingSearchResult { query: string; results: SerperShoppingResult[]; // num caps it, default 20 credits: number | null; warnings: string[]; }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/serper/searchShoppingCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.serper.searchShopping("sony wh-1000xm5");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.serper.searchShopping("sony wh-1000xm5"))curl -X POST https://api.bowmark.ai/v1/sites/serper/searchShopping \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"query":"sony wh-1000xm5"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.serper.searchShopping("sony wh-1000xm5");
Use it from code
- Node
npm i @bowmark/web - Python
pip install bowmark-web - HTTP
POST /v1/sites/serper/<function> - OpenAPI
/v1/sites/serper/openapi.json