mcpsurfr runs on your own machine — the client starts it, so there is no endpoint to ping. 778 installs a week from pypi. Last commit 29 Oct 2026.
Local web search with multi-engine deep research and on-device re-ranking. No API keys.
Today is the operative word: we check mcpsurfr every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add mcpsurfr -- uvx mcpsurfr
{
"mcpServers": {
"mcpsurfr": {
"command": "uvx",
"args": [
"mcpsurfr"
]
}
}
}
[mcp_servers.mcpsurfr]
command = "uvx"
args = ["mcpsurfr"]
{
"mcpServers": {
"mcpsurfr": {
"command": "uvx",
"args": [
"mcpsurfr"
]
}
}
}
{
"servers": {
"mcpsurfr": {
"type": "stdio",
"command": "uvx",
"args": [
"mcpsurfr"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
MCPSURFR_MAX_CHARS (Text budget per call for the full-page tools, shared across pages (~4 chars per token)).
MyWebSearch - multi-engine web search MCP server, CLI and local daemon. No API keys required.
Keyless multi-engine web search, fetch, and clean-Markdown read for AI agents. No API keys.
No-API-key web search, page fetching and document reading across many engines. Local-first.
Web search, crawling and ranking evaluation on your own YaCy peer: no API key, no central service.
Web search, page fetching and multi-source research via the DeepSeek Search & Fetch API.
Free web search for AI agents — multi-engine parallel search, smart ranking, no sign-up, no limits
Free web search and long-term memory for AI assistants. No API keys; data stays local.
Web search and web fetch with no API keys. Reads the sites that refuse AI agents.
Answers built from our own checks of this server.