Darus Search Engine
Provides social search capabilities for Bluesky, enabling discovery of posts and community content.
Provides social search capabilities for Dev.to, enabling discovery of developer articles and community posts.
Enables web and news searching through DuckDuckGo.
Provides GitHub repository and code search capabilities.
Enables news aggregation and searching from Google News.
Provides social search capabilities for Lemmy, enabling discovery of posts across Lemmy communities.
Provides social search capabilities for Medium, enabling discovery of articles and publications.
Enables web searching through Mojeek.
Provides social search capabilities for Product Hunt, enabling discovery of products and launches.
Provides social search capabilities for Reddit, enabling discovery of posts and discussions.
Provides social search capabilities for Stack Overflow, enabling discovery of questions and answers.
Enables news aggregation and searching from The Guardian.
Enables web searching through Wikipedia.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Darus Search EngineResearch the latest developments in AI"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
🔎 Darus Search Engine
Darus Search Engine adalah MCP (Model Context Protocol) untuk membantu AI melakukan pencarian dan riset web secara cepat, multi-sumber, dan terstruktur.
Project ini merupakan modifikasi dari RivalSearchMCP dengan fokus utama pada Ultra Speed: eksekusi asynchronous/concurrent, connection reuse, caching, dan pengurangan bottleneck internal sambil tetap mempertahankan coverage sumber dan kemampuan tool.
🚀 Cocok Untuk Apa?
Darus Search Engine cocok untuk AI agent, chatbot, coding agent, research assistant, dan aplikasi yang membutuhkan akses ke data web tanpa harus membangun seluruh sistem search/crawling dari nol.
Contoh penggunaan:
🔎 Pencarian web multi-engine
📰 Riset berita dan informasi terbaru
💬 Pencarian komunitas/social web
🐙 Pencarian repository dan kode GitHub
🔬 Pencarian paper akademik dan dataset
🌐 Crawling / pemetaan website
📄 Membaca dan menganalisis halaman, dokumen, PDF, DOCX, dan gambar/OCR
🧠 Riset topik dan profiling entity dengan menggabungkan banyak sumber
🔍 Mengambil, menilai, dan membandingkan konten dari beberapa URL
Related MCP server: Research Powerpack MCP
⚡ Ultra Speed
Darus mempertahankan beberapa sumber pencarian dan menjalankannya secara concurrent sehingga satu provider yang lambat tidak perlu membuat provider lain berjalan secara serial.
Optimasi mencakup:
Concurrent multi-source search
HTTP connection pooling / reuse
Cache untuk operasi yang mahal dan request berulang
Concurrent content fetching dan crawling
Fast default search tanpa otomatis membuka semua halaman hasil
Per-source failure isolation agar kegagalan satu sumber tidak menghentikan sumber lain
Catatan: Ultra Speed mengurangi bottleneck internal aplikasi. Latency dan rate limit dari provider eksternal seperti search engine, GitHub, Reddit, dan sumber web lainnya tetap berada di luar kontrol server.
🛠️ MCP Tools
Darus Search Engine menyediakan 9 tool utama:
web_search— DuckDuckGo, Bing, Yahoo, Mojeek, Wikipediasocial_search— Reddit, Hacker News, Dev.to, Product Hunt, Medium, Stack Overflow, Bluesky, Lobste.rs, Lemmynews_aggregation— Google News, Bing News, The Guardian, GDELT, DuckDuckGo Newsgithub_search— pencarian repository/kode GitHubmap_website— crawling dan pemetaan websitecontent_operations— retrieve, stream, analyze, extract, score, conflict checking, dan validasi kontenresearch_topic— riset topic atau entity lintas sumberscientific_research— paper akademik dan datasetdocument_analysis— PDF, DOCX, text, image, dan OCR
🔗 MCP Endpoint
Endpoint HTTP yang digunakan oleh deployment FastMCP:
https://darus-search-engine.fastmcp.app/mcpPenting: URL di atas adalah target endpoint deployment untuk project Darus Search Engine. Endpoint baru dapat digunakan setelah repository ini dideploy ke FastMCP/Horizon dengan slug darus-search-engine. Repository GitHub sendiri tidak otomatis membuat hostname *.fastmcp.app aktif.
Untuk self-host, konfigurasi aplikasi menggunakan HTTP MCP di port 3000 dengan path /mcp/.
📦 Open Source
Darus Search Engine dikembangkan dari source code RivalSearchMCP yang berlisensi MIT. Silakan lihat repository upstream dan pertahankan attribution/license yang berlaku saat mendistribusikan modifikasi.
Upstream: https://github.com/damionrashford/RivalSearchMCP
⚠️ Batasan
Darus dapat mengoptimalkan concurrency, caching, dan overhead internal, tetapi tidak dapat menghapus rate limit atau downtime yang diberlakukan oleh layanan eksternal. Menghapus timeout aplikasi juga berarti request upstream yang benar-benar hang dapat menunggu tanpa batas.
This server cannot be deployed
Maintenance
Related MCP Connectors
Web research for agents: quality-scored Google search, webpage extraction, and deep research.
Agent-native search engine with live web research optimized for AI agents.
Web search, fetch, extract, and research for AI agents. Markdown output + AI-synthesized answers.
The best web search for your AI Agent
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceAn agent-based tool that provides web search and advanced research capabilities including document analysis, image description, and YouTube transcript retrieval.17Apache 2.0
- FlicenseAqualityNot gradedmaintenanceEnables AI assistants to perform comprehensive research by searching Google, mining Reddit discussions, scraping web content with JS rendering, and synthesizing findings with citations into structured context.5165 npm3-
- AlicenseAqualityCmaintenanceEnables AI assistants to perform comprehensive web research through tiered search, secure URL fetching with markdown conversion, and automated multi-source synthesis pipelines. Provides read-only tools with configurable caching, SSRF protection, and optional LLM-powered summarization for search results and content analysis.81MIT
- AlicenseAqualityAmaintenanceEnables AI agents to search and fetch high-quality information from multiple sources, including general web APIs, Wikipedia, arXiv, Hacker News, Stack Exchange, and Crossref, with optional pro-mode deep research and clean markdown page extraction.10MIT