ContextBolt SEO in JetBrains IDEs (2026 Setup)

Set it up in JetBrains

Connect your ContextBolt account For OAuth, add https://seo.contextbolt.app/mcp and complete sign-in with your SEO account email and email code, then allow access. If your client cannot complete OAuth, follow the personal-URL compatibility steps on this page. Get that private URL from your ContextBolt SEO welcome email. Use one connection method.
Personal-URL setup: Open the MCP settings panel Go to Settings, then Tools, then AI Assistant, then Model Context Protocol (MCP). This panel is shared across every JetBrains IDE, so the same setup works in IntelliJ IDEA, WebStorm, PyCharm, PhpStorm, GoLand, RubyMine, and Rider.
Personal-URL setup: Add the server with its URL Click Add and paste a config with your MCP URL in the url field. JetBrains' docs show a remote server with just a url, and AI Assistant connects over Streamable HTTP, so no type field is needed. Goose and VS Code use their own type values, so do not copy those across.
Personal-URL setup: Confirm the tools and connect Search Console Apply the settings, open AI Assistant chat, and ask 'what SEO tools do you have?'. Then ask 'connect my Google Search Console' and approve read-only access to read your real clicks and positions. The Search Console tools are free and never spend your monthly credits.

The guide covers every client and the compatibility options. The full setup guide

What to ask it

“What should this page target, and how hard is it to rank for?” Chains keyword_research and keyword_difficulty so you choose a target and see the realistic cost of competing for it.

“Who ranks in the top 10 for 'headless cms comparison'?” Calls serp_overview and returns the live page-one results, so you can see the shape of what Google wants before writing.

“What does our biggest competitor rank for that we don't?” Runs competitor_keywords for the rivals, then ranked_keywords on both sides for the gap list, turning your content backlog into something data-led.

“Which of our pages are slipping in Search Console this month?” Reads your own Search Console data to flag pages losing position, free and without spending credits.

What it can do in JetBrains

Run keyword research in plain English inside any JetBrains IDE
Score how hard a keyword is to rank for before you write
See who ranks in Google's top 10 for any query
Size up any domain's traffic, keywords, and authority
Find the keywords a competitor already ranks for
Spot backlink gaps: who links to rivals but not you
Connect your own Google Search Console, read-only and free
One config that works across IntelliJ, WebStorm, PyCharm, and the rest
Get a private SEO Dashboard that fills itself with your research, free

ContextBolt SEO in JetBrains: the full guide

JetBrains IDEs were among the last major editors to get full streamable HTTP support, which meant JetBrains users spent a long time watching the MCP ecosystem from the outside. That gap has closed. AI Assistant now connects to remote MCP servers over streamable HTTP directly, so ContextBolt SEO works in IntelliJ IDEA, WebStorm, PyCharm, PhpStorm, GoLand, RubyMine, and Rider with one config entry and no bridge.

One config, every IDE

Use https://seo.contextbolt.app/mcp and complete the browser sign-in if your client supports remote MCP with OAuth. Enter your SEO account email and email code, then allow the connection. For a client that cannot complete that flow, use the personal-URL compatibility configuration shown here. Copy the full private URL from your ContextBolt SEO welcome email and replace the example URL with it.

The MCP panel lives inside AI Assistant rather than in any individual IDE, which is genuinely useful if you work across several. The same server entry lights up WebStorm for front-end work and PyCharm for the backend, and you configure it once.

Go to Settings, then Tools, then AI Assistant, then Model Context Protocol (MCP), and click Add:

{
  "mcpServers": {
    "contextbolt-seo": {
      "url": "https://seo.contextbolt.app/mcp/YOUR_TOKEN"
    }
  }
}

Because your token sits in the URL path, you do not need any headers.

No type field needed

JetBrains’ docs show a remote server as a name and a url, and say AI Assistant connects to it over the Streamable HTTP transport. Goose expects streamable_http as a type and VS Code wants http, so do not paste a type value from another tool’s config into this one.

If you have pasted a config and the tools never appear, check that the URL is complete first.

Recent releases also added auto-reconnect across all transports, so a dropped network connection no longer means re-adding the server by hand.

What this changes for backend-heavy teams

JetBrains users tend to be underserved by MCP content, because the ecosystem writes for VS Code and its forks. The practical effect is that SEO work gets done somewhere else entirely, in a dashboard, by a different person, usually after the page has shipped.

Having the tools in the IDE moves that decision earlier. You are building a marketing page in WebStorm, you ask what it should target, and you write the copy against a real query instead of a guess. When it ships, you ask for an audit and fix what comes back in the same session.

Plan before you build: keyword research and difficulty scoring, while the file is open.

Audit after you ship: point page_audit at the live URL and work the fix list.

Read your own data free: the Search Console tools never spend credits, so checking what moved this month costs nothing.

The full tour is in the ContextBolt SEO guide, and the background on how an SEO MCP server works is worth ten minutes before you set one up.

Frequently asked questions

Something else on your mind? Get in touch.

Which JetBrains IDEs does this work in?

All of them that ship AI Assistant, because the MCP panel lives in AI Assistant rather than in any one IDE. IntelliJ IDEA, WebStorm, PyCharm, PhpStorm, GoLand, RubyMine, and Rider all use the same Settings, Tools, AI Assistant, Model Context Protocol path, and the same config works in each.

Do I need a type field?

No. JetBrains' docs show a remote server config with only a url, which AI Assistant connects to over Streamable HTTP. Goose uses type streamable_http and VS Code uses type http, so a type value copied from another tool is not something to rely on here. If your server will not connect, check the URL is pasted in full first.

Do I need a paid AI Assistant subscription?

You need AI Assistant enabled in your IDE, which is where the MCP client lives. ContextBolt SEO is billed separately at $35/month for 1,000 credits. Check your current JetBrains plan, since what is included in the free tier has changed more than once.

Does JetBrains reconnect automatically if the server drops?

Yes. Recent releases added auto-reconnect across all transports, including streamable HTTP, so a brief network drop does not mean re-adding the server. If the tools stop appearing entirely, reopen the MCP panel and check the server still shows as connected.

Can it write findings to my project folder?

To your ContextBolt account. Every lookup lands on your SEO Dashboard, a private page you can sign in to from any machine. It does not depend on what file access you granted AI Assistant, because nothing is written to your disk.

How many credits does a question cost?

Research tools cost one credit each. Backlink tools cost three each. The plan includes 1,000 a month, top-ups never expire, and checking your balance is free.

Go deeper

Keep reading

All integrations