Best MCP Servers for JetBrains IDEs in 2026

JetBrains' AI Assistant picked up native MCP client support across the whole IntelliJ-platform family back in version 2025.2, and Rider's July 2026 release pushed the family's agent story further with built-in agent skills. That's a huge install base (bigger than any single client already covered in this series), and nobody's published a JetBrains-specific guide for web-data MCP servers yet. This one compares what actually connects, with a config detail that trips up a naive copy-paste.

Two different "MCP" things JetBrains ships: don't confuse them

JetBrains IDEs expose MCP in two directions, and vendor docs sometimes blur them:

1. The IDE's own built-in MCP server (Settings → Tools → MCP Server): this exposes the IDE's own capabilities (run tests, refactor, read project structure) so an external agent like Claude Code or GitHub Copilot CLI can call into the IDE. This is not what this guide is about.
2. AI Assistant's MCP client (Settings → Tools → AI Assistant → Model Context Protocol): this is what lets AI Assistant, running inside the IDE, call out to an external MCP server like Firecrawl, Bright Data, or Crawlora. This is the direction this guide covers.

Both need IDE version 2025.2 or later.

Configuring an external MCP server in AI Assistant

Settings → Tools → AI Assistant → Model Context Protocol → Add, choose a connection type (STDIO or HTTP), and paste a JSON block shaped like:

{
  "mcpServers": {
    "yourServerName": {
      "command": "npx",
      "args": ["-y", "npm-package-name"]
    }
  }
}

or, for a remote server:

{
  "mcpServers": {
    "yourServerName": {
      "url": "https://example.com/mcp"
    }
  }
}

The gotcha: that remote url form has no headers field in the UI, so there's no built-in way to attach an Authorization: Bearer <key> header the way Warp or Windsurf's config allows directly. The practical workaround (used for any Bearer-token remote MCP server, not just the ones in this guide) is to bridge it through a local STDIO command with mcp-remote, which carries the header for you:

{
  "mcpServers": {
    "crawlora": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.crawlora.net/mcp",
        "--header",
        "Authorization:${AUTH_HEADER}"
      ],
      "env": { "AUTH_HEADER": "Bearer <CRAWLORA_API_KEY>" }
    }
  }
}

OAuth-flow servers aren't supported yet either. As of this writing, JetBrains' MCP client only handles static headers and local processes.

What to evaluate

• Generic crawl (any URL → markdown) vs. structured platform data (named sites → JSON).
• Does the vendor actually document JetBrains, or are you adapting a Claude Desktop config and hoping the header quirk doesn't bite you?
• Tool count and what it actually covers.
• Pricing model and whether a failed request still gets billed.

The contenders

Firecrawl MCP: generic crawl-to-markdown, no header workaround needed

Firecrawl's MCP server runs locally via npx -y firecrawl-mcp with FIRECRAWL_API_KEY under env, a command-based entry, so JetBrains' remote-header limitation never comes into play. It scrapes, crawls, maps, searches, and extracts from any public URL, not a fixed platform catalog. Choose it when the job is reading or crawling arbitrary websites.

Bright Data: 60+ tools, no native-AI-Assistant guide, but a real path via Augment Code or Zencoder

Bright Data's core docs and GitHub README list Claude, Cursor, VS Code, Windsurf, Zed, and Warp by name but not JetBrains' own AI Assistant, and there's no dedicated setup guide for that specific path. There is real, published Bright Data content for JetBrains users, though: separate blog posts walk through connecting Bright Data's MCP server to Augment Code (whose IDE plugin, "Auggie," lists JetBrains among its supported IDEs) and to Zencoder (whose JetBrains plugin uses the same MCP config as its VS Code one). Neither is JetBrains' native AI Assistant, so if that's specifically what you're running, you're back to a generic setup: it's a command-based local process (npx @brightdata/mcp, API_TOKEN in env, PRO_MODE: "true" for the full 60+-tool set), so it sidesteps JetBrains' remote-header limitation the same way Firecrawl does, and any MCP-compatible client can point at it. Choose it when reliable extraction from specific, hard, named platforms is the actual constraint, whether you reach it through AI Assistant directly or through a JetBrains-compatible extension like Augment Code or Zencoder.

Zyte: not currently available for JetBrains

Zyte's two Claude-adjacent products, the Claude Code plugin and the GitHub-Copilot-specific Web Scraping Copilot VS Code extension, are each locked to one client. Neither connects to any JetBrains IDE as of this writing.

Crawlora: structured JSON across 155 platform groups, via the mcp-remote bridge

Crawlora's MCP server is a remote HTTP endpoint, which means it's the one vendor here that actually runs into JetBrains' header limitation. Use the mcp-remote bridge block above rather than a bare url entry, or the Authorization header silently won't be sent. Once connected: 1,398 tools across search, maps, e-commerce, social, finance, travel, jobs, media, app stores, and reviews, normalized JSON, pay-on-success billing, 2,000 free credits/month with no card. Choose it for structured records from named platforms without maintaining a parser.

How to choose, by job

• You need to crawl or read arbitrary websites for agent context → Firecrawl MCP.
• You need reliable extraction from specific hard, named platforms at scale → Bright Data, following its Augment Code or Zencoder guide if you use one of those, or the generic MCP setup otherwise.
• You need structured JSON from named platforms without maintaining a parser, billed only on success → Crawlora MCP (via the mcp-remote bridge).
• You specifically want Zyte's Scrapy-generation workflow → you'll need Claude Code or GitHub Copilot for now, not a JetBrains IDE.

Add structured web data to your JetBrains IDE

1,398 tools across 155 platform groups, one API key, billed only on success. 2,000 free credits a month, no card.

Set up Crawlora MCPBrowse the API catalog

Sources

Related reading

• Best MCP Servers for Warp in 2026: another client that recently expanded its reach, same three vendors.
• Best MCP Servers for GitHub Copilot in 2026.
• Best MCP Servers for Windsurf in 2026.
• Best Web Scraping MCP Servers & Skills for Claude Code in 2026.
• Crawlora vs Firecrawl · Crawlora vs Bright Data.
• Give your AI agent live web data with MCP.