keyless Google web SERP scrape (HTTP first, Playwright
How to use it
google is exposed by the Free Search MCP MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the google tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.
Install Free Search MCP
claude mcp add search -- uvx free-search-mcp # Claude CodeOther tools in Free Search MCP (28)
AnySearch
keyless Bilibili (哔哩哔哩) video search via the public
, baidu — intermittent challenges to headless clients.
Brave Search API
news / pdf / github / paper / forum / blog
zenodo
["pinterest.com"]
"beginner"
stackexchange, hackernews
day / week / month / year
github (repos + issues/PRs), githubcode (needs a token)
GitHub
Google Custom Search
openverse
["python.org", "djangoproject.com"]
"async"
24
googlenews, gdelt
book search over the Internet Archive catalogue.
arxiv, openalex, crossref, pubmed
meta-search proxy via public SearXNG instances; included for
Serper (Google)
, so360 — Chinese web indexes, HTML scrapes, best-effort like
Stack Apps (key optional)
browser-rendered (~5-10s/query); good for hard-to-reach
Tavily (AI search)
encyclopedia search; language follows SEARCHMCPREGION.
best-effort keyless Zhihu (知乎) search. Zhihu's