MCP servers that search the web
Give your AI live web search results with sources.
28 servers · Last updated August 1, 2026
TL;DR: These servers return fresh search results so your agent can answer with current information and cite sources, instead of relying on stale training data. They differ on index (general vs developer-focused), whether they return full page content or just snippets, and pricing model.
Bottom line: if you only try one, Firecrawl is the most popular, verified option for this (6,562★). 27 more compared below.
Compare 28 servers
| Server | Transport | Auth | Verified | Stars | Tools for this |
|---|---|---|---|---|---|
| Firecrawl | Local (stdio) | API key | 6,562 | firecrawl_search | |
| Exa Search | Local (stdio) | API key | 4,573 | web_search_exa, web_search_advanced_exa, company_research_exa +1 | |
| ArXiv MCP Server | Local (stdio) | No auth | — | 2,900 | search_papers, semantic_search |
| Perplexity | Local (stdio) | API key | 2,299 | perplexity_search, perplexity_research | |
| Tavily | Local (stdio) | API key | 2,100 | tavily-search | |
| Apify Actors | Local (stdio) | API key | 1,336 | search-actors, search-apify-docs | |
| DuckDuckGo Search | Local (stdio) | No auth | 1,236 | search | |
| Brave Search | Local (stdio) | API key | 1,179 | brave_web_search, brave_local_search, brave_image_search +3 | |
| Ref MCP | Local (stdio) | API key | — | 1,134 | ref_search_documentation |
| Web Search MCP Server | Local (stdio) | No auth | — | 980 | full-web-search, get-web-search-summaries |
| SearXNG | Local (stdio) | No auth | 896 | searxng_web_search, searxng_search_suggestions | |
| Jina AI Reader & Search | Remote (HTTP) | API key | 730 | search_web, search_images, search_arxiv +1 | |
| BioMCP | Local (stdio) | No auth | — | 527 | search <entity> [filters], search all [slot filters] |
| Web Search MCP Server | Local (stdio) | No auth | — | 452 | search |
| Kagi Search | Local (stdio) | API key | 417 | kagi_search_fetch | |
| Mapbox MCP Server | Local (stdio) | API key | — | 347 | search_and_geocode_tool, category_search_tool |
| Everything Search MCP Server | Local (stdio) | No auth | — | 336 | search |
| Elasticsearch/OpenSearch MCP Server | Local (stdio) | API key | — | 290 | search_documents |
| Wikipedia MCP | Local (stdio) | No auth | — | 270 | search_wikipedia |
| Facebook Ads Library MCP Server | Local (stdio) | API key | — | 243 | search_cached_media |
| Meilisearch MCP Server | Local (stdio) | API key | — | 192 | search |
| Bilibili MCP | Local (stdio) | No auth | — | 174 | search-video |
| Discogs MCP Server | Local (stdio) | API key | — | 112 | search |
| WebSearch MCP | Local (stdio) | No auth | — | 37 | web_search |
| Met Museum MCP Server | Local (stdio) | No auth | — | 30 | search-museum-objects |
| Decodo MCP Server | Remote (HTTP) | API key | — | 30 | google_search, amazon_search, walmart_search +4 |
| Algolia MCP Server | Local (stdio) | API key | — | 28 | search, search_read, search_write |
| Inkeep MCP Server | Local (stdio) | API key | — | 25 | search-product-content |
The servers
Official Firecrawl MCP server — scrape, crawl, map, search, and structured extraction for any LLM client.
Exa's neural web search and crawling MCP server — runs locally via npx or as a hosted remote endpoint.
Search, download, and read arXiv research papers from your AI assistant via MCP.
Official Perplexity MCP server — web search, ask, deep research, and reasoning via Sonar models.
Production-ready MCP server for real-time web search, content extraction, site mapping, and crawling.
Official Apify MCP server — search and scrape the web via thousands of ready-made Actors; local or remote.
Popular no-API-key MCP server for DuckDuckGo web search plus page fetching and parsing.
Official Brave Search MCP server: web, local, image, video, news search and AI summarization.
Token-efficient documentation search and reading for coding agents across public and private docs.
Locally hosted multi-engine web search and page content extraction for local LLMs — no API keys required.
Private metasearch via your own SearXNG instance — popular community MCP server for self-hosted search.
Official Jina AI remote MCP server — read web pages as Markdown and run grounded web search over HTTP.
One command grammar to search and pivot across PubMed, ClinVar, ClinicalTrials.gov, and dozens of biomedical sources.
Free web search via Google results scraping — no API keys required.
Official Kagi MCP server (Python/uvx) — privacy-first web search and URL/video summarization.
Geospatial intelligence for AI: geocoding, POI search, routing, isochrones, matrices, and static maps via Mapbox APIs.
Fast cross-platform file search via Everything SDK (Windows), mdfind (macOS), and locate (Linux).
Search, analyze, and manage Elasticsearch & OpenSearch clusters through MCP tools.
Give your LLM live Wikipedia access: search, full articles, summaries, sections, links and key facts in any language.
Search and analyze any brand's public Facebook ads — images, videos, messaging, and competitor strategy.
Manage and search Meilisearch from any MCP client: indices, documents, settings, keys, and multi-index search.
MCP server to search Bilibili videos and fetch trending lists, video details, UP-host info, and anime schedules.
Search and manage your Discogs music catalog, collection, wantlist, and marketplace from your AI client.
Real-time web search for MCP clients via a self-hosted crawler API over stdio.
Search and explore The Metropolitan Museum of Art collection, including artwork details and images, via natural language.
Scrape websites, search engines, eCommerce, and social media for AI agents via Decodo's Web Scraping API.
Query and manage Algolia search indices from an agent: search, settings, analytics, A/B tests, and Recommend.
Search your product docs and content via Inkeep's RAG, exposed as an MCP tool.
Use these in a stack
FAQ
Which web-search MCP is best for AI agents?
For citations and clean snippets, API-first engines (Brave, Tavily, Exa) work well; Exa skews developer/semantic. Compare them head-to-head on the directory before committing.
Do these need an API key?
Most do — you'll provision a key from the search provider and drop it in the config. A few self-hosted options (SearXNG) avoid that.