Open Web Search
๐ ๐ โ๏ธ - Web search using free multi-engine search (NO API KEYS REQUIRED) โ Supports Bing, Baidu, DuckDuckGo, Brave, Exa, and CSDN.
What is Open Web Search?
Open Web Search is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to ๐ ๐ โ๏ธ - web search using free multi-engine search (no api keys required) โ supports bing, baidu, duckduckgo, brave, exa, and csdn.
๐ ๐ โ๏ธ - Web search using free multi-engine search (NO API KEYS REQUIRED) โ Supports Bing, Baidu, DuckDuckGo, Brave, Exa, and CSDN.
This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- ๐ ๐ โ๏ธ - Web search using free multi-engine search (NO API K
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx open-websearchConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Open Web Search
Open Web Search is a zero-configuration MCP server that provides web search through multiple free search engines โ Bing, DuckDuckGo, Brave, Baidu, Exa, CSDN, Juejin, Startpage, and Sogou โ without requiring any API keys. Beyond basic search, it exposes dedicated tools for fetching full article content from CSDN, Juejin, LinuxDo forums, and arbitrary public web pages, as well as pulling GitHub README files. It runs as a standalone stdio process or an HTTP server and is designed for situations where you want AI-powered web research without the cost or signup friction of commercial search APIs.
Prerequisites
- Node.js 16 or later (for npx)
- No API keys required โ all supported engines are free to query
- An MCP-compatible client such as Claude Desktop, Cursor, or another MCP host
- Optional: Docker if you prefer a containerised deployment
Test the server locally with npx
Run the server once via npx to confirm it starts correctly before adding it to your MCP client config.
npx open-websearch@latestAdd the server to your MCP client config
Add the following entry to your claude_desktop_config.json (or equivalent MCP host config). No env vars are required for basic operation.
{
"mcpServers": {
"open-websearch": {
"command": "npx",
"args": ["open-websearch@latest"]
}
}
}Optional: set a default search engine
Use the DEFAULT_SEARCH_ENGINE environment variable to pin a preferred engine. Valid values include bing, duckduckgo, brave, baidu, exa, csdn, juejin, startpage, and sogou.
{
"mcpServers": {
"open-websearch": {
"command": "npx",
"args": ["open-websearch@latest"],
"env": {
"DEFAULT_SEARCH_ENGINE": "duckduckgo",
"ALLOWED_SEARCH_ENGINES": "bing,duckduckgo,brave"
}
}
}
}Optional: enable a proxy
If search engines are rate-limiting or blocked in your region, enable proxy support by setting USE_PROXY and PROXY_URL.
"env": {
"USE_PROXY": "true",
"PROXY_URL": "http://127.0.0.1:7890"
}Restart your MCP client and verify tools appear
Restart the MCP client. The server exposes tools: search, fetchWebContent, fetchGithubReadme, fetchCsdnArticle, fetchJuejinArticle, and fetchLinuxDoArticle. Ask the AI to search for something to confirm it is working.
Open Web Search Examples
Client configuration
Minimal claude_desktop_config.json with DuckDuckGo as the default engine and no API keys needed.
{
"mcpServers": {
"open-websearch": {
"command": "npx",
"args": ["open-websearch@latest"],
"env": {
"DEFAULT_SEARCH_ENGINE": "duckduckgo"
}
}
}
}Prompts to try
Use these prompts to exercise the search and content-fetching tools.
- "Search for the latest news on Model Context Protocol using DuckDuckGo"
- "Fetch the full content of this web page: https://example.com/article"
- "Search for 'Python async tutorial' across Bing and Brave and compare the top results"
- "Fetch the README from https://github.com/anthropics/anthropic-sdk-python"
- "Search CSDN for articles about Spring Boot performance tuning"Troubleshooting Open Web Search
Search returns no results or times out
Try a different engine by specifying it in the search tool call (e.g. engines: ["brave","bing"]). Some engines may be rate-limited by IP. Enable USE_PROXY and set PROXY_URL to route through a different network address.
fetchWebContent fails on a page that exists
Some pages require JavaScript rendering. Set SEARCH_MODE to 'playwright' to enable Playwright-based headless browser fetching, though this requires Playwright to be installed on the host system.
Server starts but no MCP tools are listed in the client
Confirm you are running a recent version via `npx open-websearch@latest --version`. If using a locked version, update to latest. Also check that MODE is set to 'stdio' (the default) rather than 'http' when using an MCP client that connects over stdio.
Frequently Asked Questions about Open Web Search
What is Open Web Search?
Open Web Search is a Model Context Protocol (MCP) server that ๐ ๐ โ๏ธ - web search using free multi-engine search (no api keys required) โ supports bing, baidu, duckduckgo, brave, exa, and csdn. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Open Web Search?
Follow the installation instructions on the Open Web Search GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Open Web Search?
Open Web Search works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Open Web Search free to use?
Yes, Open Web Search is open source and available under the Apache 2.0 license. You can use it freely in both personal and commercial projects.
Open Web Search Alternatives โ Similar Search & Data Extraction Servers
Looking for alternatives to Open Web Search? Here are other popular search & data extraction servers you can use with Claude, Cursor, and VS Code.
TrendRadar
โ 58.0kA real-time hotspot monitoring and news aggregation assistant that provides AI-powered analysis of trending topics across multiple platforms via the Model Context Protocol. It enables users to track news and receive automated notifications through va
Scrapling
โ 52.7k๐ท๏ธ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!
PDF Math Translate
โ 33.9k[EMNLP 2025 Demo] PDF scientific paper translation with preserved formats - ๅบไบ AI ๅฎๆดไฟ็ๆ็็ PDF ๆๆกฃๅ จๆๅ่ฏญ็ฟป่ฏ๏ผๆฏๆ Google/DeepL/Ollama/OpenAI ็ญๆๅก๏ผๆไพ CLI/GUI/MCP/Docker/Zotero
GPT Researcher
โ 27.2kAn autonomous agent that conducts deep research on any data using any LLM providers
Agent Reach
โ 20.1kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu โ one CLI, zero API fees.
Xiaohongshu
โ 13.7kMCP for xiaohongshu.com
Browse More Search & Data Extraction MCP Servers
Explore all search & data extraction servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Open Web Search in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Open Web Search?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.