MCP server · Exa

exa

Exa operates an official remote MCP server at https://mcp.exa.ai/mcp. It runs a free, rate-limited tier with no auth; paid/agent tools authenticate with an Exa API key (x-api-key header) or OAuth. The OAuth protected-resource metadata (well-known/exa-ai-mcp-oauth-protected-resource.json) points at the authorization server https://auth.exa.ai (scope mcp:tools, PKCE S256). A local stdio server ships as npx exa-mcp-server (github.com/exa-labs/exa-mcp-server); a dedicated Websets MCP server is at github.com/exa-labs/websets-mcp-server. Tools captured verbatim from the Exa MCP documentation.

Open endpoint → Exa provider profile

One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.

Provider: Exa Type: Hosted endpoint Transport: http Host: mcp.exa.ai

Endpoint

https://mcp.exa.ai/mcp

Hosted endpoint · transport http

Connect

This is a remote MCP server — point an MCP client at the endpoint URL. Clients with native remote support (Claude, Cursor, VS Code, …):

# mcp client config (e.g. .mcp.json / mcp settings) { "mcpServers": { "exa-ai": { "url": "https://mcp.exa.ai/mcp" } } }

For stdio-only clients (older Claude Desktop, etc.), bridge with mcp-remote:

# Add to claude_desktop_config.json { "mcpServers": { "exa-ai": { "command": "npx", "args": ["-y", "mcp-remote", "https://mcp.exa.ai/mcp"] } } }

If the server requires authentication, add the provider's token/header (e.g. an X-API-Key) per its docs. A quick reachability check:

curl -i -X POST https://mcp.exa.ai/mcp \ -H "Content-Type: application/json" -H "Accept: application/json, text/event-stream" \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

Tools

web_search_exa — Search the web for any topic and get clean, ready-to-use content.
web_fetch_exa — Read a webpage's full content as clean markdown from one or more URLs.
get_code_context_exa — Search for code and technical documentation context.
web_search_advanced_exa — Advanced web search with category filters, domain restrictions, date ranges, highlights, summaries, and subpage crawling.
crawling_exa — Crawl and extract content from specific URLs.
company_research_exa — Research companies with structured entity data.
linkedin_search_exa — Search LinkedIn profiles and companies.
deep_researcher_start — Start an asynchronous deep-research task.
deep_researcher_check — Poll a deep-research task for completion and results.
agent_create_run — Start async Agent runs for multi-step research, list-building, enrichment, or structured JSON output.
agent_wait_for_run — Poll a run until terminal status or timeout.
agent_get_run_output — Retrieve completed output with text, structured results, citations, usage, and cost data.
agent_cancel_run — Stop a queued or running operation.

About MCP

The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.

Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.