MetaSearch
Open-source metasearch backend, MCP server, and AI search API for LLM agents. Python FastAPI search gateway with Google search via SerpBase and Serper, multi-engine search aggregation, structured JSON output, provider fallback, deduplication, and Sea
What is MetaSearch?
MetaSearch is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to open-source metasearch backend, mcp server, and ai search api for llm agents. python fastapi search gateway with google search via serpbase and serper, multi-engine search aggregation, structured json...
Open-source metasearch backend, MCP server, and AI search API for LLM agents. Python FastAPI search gateway with Google search via SerpBase and Serper, multi-engine search aggregation, structured JSON output, provider fallback, deduplication, and Sea
This server falls under the Search & Data Extraction category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Open-source metasearch backend, MCP server, and AI search AP
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx metasearchmcpConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MetaSearch
MetaSearchMCP is an open-source metasearch gateway that aggregates results from multiple search engines and data sources — Google (via SerpBase or Serper), Brave, GitHub, academic databases, Reddit, finance APIs, and more — into a single structured JSON response for LLM agents. It runs as both a FastAPI HTTP server and an MCP server, exposing seven search tools with automatic provider fallback, deduplication, and configurable result limits. This makes it an ideal drop-in search backend for AI agents that need reliable, multi-source search without being locked into a single provider.
Prerequisites
- Python 3.10+ with pip installed
- At least one search provider API key: SERPBASE_API_KEY or SERPER_API_KEY for Google results, BRAVE_API_KEY for web search
- Optional keys for extended providers: GITHUB_TOKEN, ALPHA_VANTAGE_API_KEY, NCBI_API_KEY, etc.
- An MCP client such as Claude Desktop, or an HTTP client for the FastAPI endpoint
- Git to clone the repository
Clone the repository
Clone MetaSearchMCP from GitHub and enter the project directory.
git clone https://github.com/gefsikatsinelou/MetaSearchMCP.git
cd MetaSearchMCPRun the automated installer
The install script sets up a virtual environment, installs dependencies, and creates a .env file from the template.
python scripts/install.pyConfigure your API keys
Edit the .env file created by the installer and add at least one search provider API key. Multiple keys enable fallback behavior.
# .env
SERPBASE_API_KEY=your_serpbase_key
SERPER_API_KEY=your_serper_key
BRAVE_API_KEY=your_brave_key
GITHUB_TOKEN=your_github_token
ALPHA_VANTAGE_API_KEY=your_alpha_vantage_key
ENABLED_PROVIDERS=serpbase,serper,brave,github
MAX_RESULTS_PER_PROVIDER=10Add to your MCP client configuration
Register the MetaSearchMCP server in your claude_desktop_config.json using the metasearchmcp-mcp command.
{
"mcpServers": {
"MetaSearchMCP": {
"command": "metasearchmcp-mcp",
"env": {
"SERPBASE_API_KEY": "your_serpbase_key",
"BRAVE_API_KEY": "your_brave_key",
"ALLOW_UNSTABLE_PROVIDERS": "true"
}
}
}
}Restart and verify the connection
Restart your MCP client and confirm the MetaSearchMCP server is listed as connected. The server exposes seven tools: search_web, search_google, search_academic, search_github, compare_engines, search_finance, and search_code.
MetaSearch Examples
Client configuration
Full claude_desktop_config.json entry for MetaSearchMCP with Google and Brave providers.
{
"mcpServers": {
"MetaSearchMCP": {
"command": "metasearchmcp-mcp",
"env": {
"SERPBASE_API_KEY": "your_serpbase_api_key",
"SERPER_API_KEY": "your_serper_api_key",
"BRAVE_API_KEY": "your_brave_api_key",
"GITHUB_TOKEN": "your_github_token",
"ALLOW_UNSTABLE_PROVIDERS": "true",
"MAX_RESULTS_PER_PROVIDER": "10"
}
}
}
}Prompts to try
Sample prompts to use with MetaSearchMCP connected in your AI assistant.
- "Search the web for 'Rust async runtime comparison 2025' using all available providers"
- "Search GitHub for repositories about 'MCP server Python' with max 20 results"
- "Search academic sources for papers on 'large language model hallucination mitigation'"
- "Compare search engine results for 'best vector database 2025' across Google and Brave"
- "Search finance data for Apple stock news and analyst reports"
- "Search code repositories for examples of 'FastAPI streaming responses'"Troubleshooting MetaSearch
All search tools return empty results or provider errors
Check that at least one provider API key is set correctly in the .env file or environment. Run 'python scripts/install.py --test' to validate provider connectivity before starting the MCP server.
metasearchmcp-mcp command not found after installation
Ensure the package was installed in editable mode with 'pip install -e .[dev]' inside the project directory. Alternatively, run the server directly with 'python -m metasearchmcp.mcp_server' and adjust the MCP config command accordingly.
Rate limit errors from a specific provider
Set ALLOW_UNSTABLE_PROVIDERS=true to enable provider fallback. The server will automatically skip rate-limited providers and use the next available one. Reduce MAX_RESULTS_PER_PROVIDER to lower API usage.
Frequently Asked Questions about MetaSearch
What is MetaSearch?
MetaSearch is a Model Context Protocol (MCP) server that open-source metasearch backend, mcp server, and ai search api for llm agents. python fastapi search gateway with google search via serpbase and serper, multi-engine search aggregation, structured json output, provider fallback, deduplication, and sea It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MetaSearch?
Follow the installation instructions on the MetaSearch GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with MetaSearch?
MetaSearch works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is MetaSearch free to use?
Yes, MetaSearch is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
MetaSearch Alternatives — Similar Search & Data Extraction Servers
Looking for alternatives to MetaSearch? Here are other popular search & data extraction servers you can use with Claude, Cursor, and VS Code.
TrendRadar
★ 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
★ 52.7k🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!
PDF Math Translate
★ 33.9k[EMNLP 2025 Demo] PDF scientific paper translation with preserved formats - 基于 AI 完整保留排版的 PDF 文档全文双语翻译，支持 Google/DeepL/Ollama/OpenAI 等服务，提供 CLI/GUI/MCP/Docker/Zotero
GPT Researcher
★ 27.2kAn autonomous agent that conducts deep research on any data using any LLM providers
Agent Reach
★ 20.1kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
Xiaohongshu
★ 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 MetaSearch 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 MetaSearch?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.