MCP server · Web + search

OfficialSource ValidatedRead-onlyMITUpdated August 2026

Exa MCP

Search the web and fetch clean page content through Exa, with optional advanced search and multi-step research tools.

InstallSource repository

Quick verdict

A clean read-oriented complement to crawler and browser MCPs, especially for quick search and fetch workflows, with a sensible anonymous on-ramp and clear paid expansion path.

  • MaintainerExa Labs (official)
  • Security scopeRead-only
  • Transportsstreamable-http, stdio
  • Install channelremote · https://mcp.exa.ai/mcp
  • EvidenceSource Validated
  • LicenseMIT · Open source

Exa MCP connects an assistant to Exa's hosted search and content-retrieval service. The anonymous starter endpoint exposes web search and page fetching without local installation; users can authenticate to raise limits and enable advanced search or Exa Agent research when their task and budget justify the wider surface.

MCP security best practices

Install Exa MCP

http https://mcp.exa.ai/mcp

Config entry (mcpServers)
{
  "mcpServers": {
    "exa": {
      "url": "https://mcp.exa.ai/mcp"
    }
  }
}

Exa-hosted endpoint; works anonymously at low rate limits with no credential. OAuth or EXA_API_KEY raises limits and unlocks Agent tools. The MIT-licensed npm wrapper can instead run locally over stdio, still calling Exa's hosted service — no self-hosted Exa search backend is documented.

Credentials

No credentials required.

Verified configurations

Not run by us yet. This profile is Source Validated: the install channel, config entry, credentials and license were checked against the source and upstream metadata, but no configuration has been installed, discovered and executed in a recorded environment. When one is, it appears here with the eight scope dimensions below.

Best for

Research and coding agents that need concise web results or clean page content without operating a browser or scraping pages directly.

Why use it

It provides an official, low-friction hosted endpoint with search and fetch tools that return agent-ready content and can be narrowed to selected capabilities.

Tradeoffs

Anonymous use is rate-limited, authenticated usage can incur per-request costs, and optional Agent tools make multiple remote decisions. Retrieved pages remain untrusted input and can contain prompt injection or stale claims. Keep source URLs, restrict enabled tools, and verify consequential facts against primary sources.

Upstream state

repository active · in the official MCP Registry (active). Lifecycle is derived from a committed upstream snapshot at build time, never authored by hand.

Official MCP Registry name: ai.exa/exa

Editorial review

Reviewed byOpenSourcesAI EditorialLast updatedAugust 2026SourcesOfficial Exa MCP documentation; Official Exa MCP source repository; Exa pricing; Exa MCP license; Official MCP Registry — Exa

Grade B — Source Validated. Re-review is due by 2027-02-18. Config entries, credential names and package status can change upstream; verify against the source repository before granting an agent access.

Sources

  • FetchReference MCP server that fetches a web page and returns it as Markdown, so a local model can read live content without a browser.
  • Firecrawl MCPSearch, scrape, parse and crawl the live web through Firecrawl, with optional browser interaction and autonomous research.
  • Tavily MCPSearch, extract, map and crawl live web content through Tavily's official hosted MCP server.

Related guides

All MCP servers →