sydasif/web-search-mcp
Developer toolsこのページは AI により自動翻訳されています。技術的な詳細は原文をご確認ください。 原文を見る
A comprehensive, production-ready research server (MCP). Provide your LLM clients with real-time access to the web, data, and more.
概要
A comprehensive server built with that provides LLMs with real-time, high-fidelity access to the web. This server aggregates multiple search engines, social platforms, and developer tools into a single interface, allowing AI agents to perform deep research, track community sentiment, and analyze technical documentation. — tool selection guide, decision matrix, recommended workflows, tools status & known quirks, plugin setup, and development standards. The server provides a diverse suite of tools categorized by their primary use case: You have three options depending on your use case: Fastest way to try it out without cloning the repo. Add to your MCP client config: Fastest startup times with a globally installed tool: If you want to modify the code or contribute: Once the server is running, try a simple search:
README
以下は原文 README の抜粋です。原文を見る。
OmniSearch
A comprehensive Model Context Protocol (MCP) server built with FastMCP that provides LLMs with real-time, high-fidelity access to the web. This server aggregates multiple search engines, social platforms, and developer tools into a single interface, allowing AI agents to perform deep research, track community sentiment, and analyze technical documentation.
Design docs → wiki — tool selection guide, decision matrix, recommended workflows, tools status & known quirks, plugin setup, and development standards.
🚀 Features
The server provides a diverse suite of tools categorized by their primary use case:
🌐 General Web Search & Retrieval
| Tool | Description | Best For |
|---|---|---|
search_web |
Fast web search via DuckDuckGo or Exa (SDK). Supports domain-scoping, date filtering, news mode, and geographic region. Default auto-provider tries DDG first, falls back to Exa. | Quick lookups, high-volume searches, pagination, broad coverage |
fetch_page |
High-fidelity text extraction from URLs with bot-detection bypass, SSRF protection (blocks private/internal IPs), and multiple output formats. | Deep reading of search results, standalone URL fetching |
💬 Social & Community Intelligence
| Tool | Description | Best For |
|---|
インストール
uvx --from git+https://github.com/sydasif/web-search-mcp.git omnisearch設定
{
"mcpServers": {
"OmniSearch": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/sydasif/web-search-mcp.git",
"omnisearch"
]
}
}
}