google_search
Tool to perform web searches via Serper API and retrieve rich results. It is able to retrieve organic search results, people also ask, related searches, and knowledge graph.
LABSCO SUMMARY
Search and scrape together is the pairing that matters, because a search result is a link and a link is not an answer.
Plenty of servers do one or the other, and the gap between them is where the work sits: finding the page, then getting its content in a form worth reading. Having both behind one key means the sequence is two calls rather than two integrations. Optional markdown on the scrape is the detail worth using — raw text loses the structure that tells a model what is a heading and what is a caption, and markdown preserves it at no extra cost.
What it is
An MCP server for the Serper API, published on npm as serper-search-scrape-mcp-server, with two tools: web search returning rich results, and page scraping returning text with optional markdown.
What you get
- google_search performs the search and returns rich results rather than a bare link list.
- scrape fetches a page and returns its text, with markdown available as an option — markdown being the form a model reads best.
Requirements
SERPER_API_KEY.
Setup effort
One command plus a key — env SERPER_API_KEY=<YOUR_SERPER_API_KEY> npx -y serper-search-scrape-mcp-server
USE CASES
01
Finding out what is on the web about something
Search and get rich results rather than a list of bare links to guess between.
02
Actually reading the page you found
Scrape it for text, taking markdown so the structure survives into what the model reads.
2 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
scrape
Tool to scrape a webpage and retrieve the text and, optionally, the markdown content. It will retrieve also the JSON-LD metadata and the head metadata.
BEFORE YOU INSTALL
What to know first
Tools you get
2 tools when we connected. google_search takes operators such as site, filetype and intitle as parameters. The project's Cline example pre-approves both tools with autoApprove, so scraping will not prompt you if copied as written.
Requires
- npx on PATH
- Node.js >= 18
- SERPER_API_KEY in your environment
CONTEXT COST
Light Standing cost
Rated by the size of the 2 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
TERMINAL
env SERPER_API_KEY=<YOUR_SERPER_API_KEY> npx -y serper-search-scrape-mcp-serverMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"serper-search": {
"command": "npx",
"args": [
"-y",
"serper-search-scrape-mcp-server"
],
"env": {
"SERPER_API_KEY": "<YOUR_SERPER_API_KEY>"
}
}
}
}Cline
CLINE · JSON
{
"mcpServers": {
"github.com/marcopesani/mcp-server-serper": {
"command": "npx",
"args": [
"-y",
"serper-search-scrape-mcp-server"
],
"env": {
"SERPER_API_KEY": "<YOUR_SERPER_API_KEY>"
},
"disabled": false,
"autoApprove": [
"google_search",
"scrape"
]
}
}
}Claude Desktop — Smithery
CLAUDE DESKTOP — SMITHERY · TERMINAL
npx -y @smithery/cli install @marcopesani/mcp-server-serper --client claude
