Scrapingdog web-scraping & SERP APIs: web scrape, Google, Amazon, LinkedIn, YouTube, and more.
A Model Context Protocol server that exposes the Scrapingdog web-scraping and SERP APIs as tools any MCP-compatible client (Claude Desktop, Claude Code, Cursor, etc.) can call.
One tool per Scrapingdog product — general web scraping, the full Google family, Amazon, Walmart, eBay, LinkedIn, X, YouTube, Bing, Baidu, and screenshots.
npm install
npm run build
The server reads your API key from the SCRAPINGDOG_API_KEY environment
variable. It is never passed as a tool argument, so it stays out of the model's
context, transcripts, and logs.
| Variable | Required | Default | Description |
|---|---|---|---|
SCRAPINGDOG_API_KEY | yes | — | Your Scrapingdog API key. |
SCRAPINGDOG_API_BASE | no | https://api.scrapingdog.com | Override the API base URL. |
SCRAPINGDOG_TIMEOUT_MS | no | 90000 | Per-request timeout in milliseconds. |
Add to claude_desktop_config.json
(macOS: ~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"scrapingdog": {
"command": "node",
"args": ["/absolute/path/to/ScrapingdogMCP/dist/index.js"],
"env": {
"SCRAPINGDOG_API_KEY": "your_api_key_here"
}
}
}
}
claude mcp add scrapingdog -e SCRAPINGDOG_API_KEY=your_api_key_here -- node /absolute/path/to/ScrapingdogMCP/dist/index.js
35 tools. Each tool's full parameter set (with descriptions and API
defaults) is generated from src/endpoints.ts and surfaced
to the client as the tool's input schema. All paths were validated against the
live API except amazon_reviews (⚠️) — see Known issues.
| Tool | Scrapingdog endpoint | Key required args |
|---|---|---|
web_scrape | /scrape | url |
| Tool | Scrapingdog endpoint | Key required args |
|---|---|---|
google_search | /google | query |
google_maps | /google_maps | query |
google_news | /google_news/v2 | — (query or a *_token) |
google_trends | /google_trends | query |
google_shopping | /google_shopping | query |
google_scholar | /google_scholar | query (or cites) |
google_jobs | /google_jobs | query |
google_finance | /google_finance | query |
google_lens | /google_lens | url |
google_ai_mode | /google/ai_mode | query |
google_ai_overview | /google/ai_overview | url |
google_shorts | /google_shorts | query |
google_hotels | /google_hotels | query, check_in_date, check_out_date |
google_patents | /google_patents | query |
google_immersive_product | /google_immersive_product | page_token |
google_images | /google_images | query |
| Tool | Scrapingdog endpoint | Key required args |
|---|---|---|
amazon_search | /amazon/search | query, domain, page, country |
amazon_product | /amazon/product | asin |
amazon_reviews ⚠️ | /amazon/reviews | asin |
amazon_offers | /amazon/offers | asin |
walmart_search | /walmart/search | url |
ebay_search | /ebay/search | url |
| Tool | Scrapingdog endpoint | Key required args |
|---|---|---|
linkedin_profile | /linkedin | type, linkId |
x_profile | /x/profile | profileId |
youtube_search | /youtube | search_query |
youtube_transcripts | /youtube/transcripts | v |
youtube_video | /youtube/video | v |
youtube_channel | /youtube/channel | channel_id |
youtube_comments | /youtube/comments | v |
| Tool | Scrapingdog endpoint | Key required args |
|---|---|---|
bing_search | /bing/search | query |
baidu_search | /baidu/search | query |
universal_search | /search | query |
| Tool | Scrapingdog endpoint | Key required args |
|---|---|---|
chatgpt | /chatgpt | prompt |
screenshot | /screenshot | url (returns an image) |
All 35 endpoints were validated against the live API and return data, with one exception:
amazon_reviews — the /amazon/reviews route is correct, but the live
endpoint currently responds HTTP 400 "Something went wrong" for every input
(any ASIN, with or without extra params). This looks like a Scrapingdog
backend/plan-scope issue rather than a wrapper bug. It stays marked with
verify: true in src/endpoints.ts until it returns data. List flagged
endpoints anytime with:npm run list:unverified
Every endpoint is a single declarative object in
src/endpoints.ts — no handler code to write. To add a
Scrapingdog product (e.g. Google Hotels, Patents, TikTok), append:
{
tool: "google_hotels",
title: "Google Hotels API",
path: "/google_hotels",
description: "…",
params: [
{ name: "query", type: "string", required: true, description: "…" },
// …
],
}
Rebuild with npm run build.
src/
types.ts Endpoint / EndpointParam type definitions
endpoints.ts Declarative registry of every Scrapingdog API
client.ts HTTP client: URL building, API-key resolution, error handling
server.ts Compiles each endpoint into an MCP tool + Zod input schema
index.ts stdio entry point
MIT
Source-derived launch command. Check the maintainer’s required arguments and credentials before running:
npx -y scrapingdog-mcpMerge this template into ~/Library/Application Support/Claude/claude_desktop_config.json. Keep existing servers. Add any arguments, credentials, and permissions required by the maintainer; this template has not been install-tested.
{
"mcpServers": {
"io-github-darshan972-scrapingdog-mcp": {
"command": "npx",
"args": [
"-y",
"scrapingdog-mcp"
]
}
}
}Restart Claude Desktop completely for changes to take effect. Confirm the server appears connected in the client’s tool list, then try a read-only example from its documentation.
Claude Desktop setup referenceScrapingdog works with any MCP-compatible client. Copy the config snippet from the Configuration section above and add it to the file shown for your client, then restart the application.
~/Library/Application Support/Claude/claude_desktop_config.jsonRestart Claude Desktop completely for changes to take effect.~/.cursor/mcp.jsonRestart Cursor for changes to take effect..vscode/mcp.jsonReload VS Code window for changes to take effect.~/.codeium/windsurf/mcp_config.jsonRestart Windsurf for changes to take effect..mcp.jsonSave at the project root, then start Claude Code in that project and review the MCP server approval prompt. Keep real credentials out of shared files.