Claude Code in JetBrains IDEs: install the plugin, features and fixes

The Claude Code JetBrains plugin connects the claude command line tool to IntelliJ IDEA, PyCharm, WebStorm and other JetBrains IDEs, so Claude's edits open in the IDE diff viewer and your current selection is shared automatically. Install the Claude Code CLI first, then install the plugin from the JetBrains Marketplace and restart the IDE.

Unlike the VS Code extension, the JetBrains plugin is not a separate chat panel. It wraps the terminal version of Claude Code and adds IDE features around it. Everything below comes from the official JetBrains IDEs page in the Claude Code docs, read on 24 October 2026.

Which JetBrains IDEs does Claude Code support?

The docs say the plugin "works with most JetBrains IDEs," and list these by name:

• IntelliJ IDEA
• PyCharm
• Android Studio
• WebStorm
• PhpStorm
• GoLand

The plugin is listed on the Marketplace as Claude Code [Beta], and its settings page in the IDE carries the same Beta label.

How do you install Claude Code in a JetBrains IDE?

There are two pieces, and you need both. The plugin "runs the claude command in your IDE's integrated terminal and connects to it. It does not bundle its own copy of the CLI."

1. Install the CLI. Follow the Claude Code quickstart or the Claude Code setup guide. If claude is not on your PATH, the plugin shows a "Cannot launch Claude Code" notification.
2. Install the plugin. Open the Marketplace in your IDE, search for Claude Code, install it and restart the IDE.
3. Log in. Run claude once. You are prompted to log in on the first run.

You need a paid Claude plan (Pro, Max, Team or Enterprise) or a Claude Console account. No API key is required for the plugin itself.

If the CLI lives somewhere the IDE can't find, open Settings > Tools > Claude Code [Beta] and set the full path in the Claude command field. The docs give claude, /usr/local/bin/claude and npx @anthropic-ai/claude-code as example values.

What does the JetBrains plugin add to Claude Code?

Once the CLI and plugin are connected, you get these features:

On diagnostics, the docs are specific: Claude reads them by calling a getDiagnostics tool, and "Claude Code doesn't request diagnostics from the plugin on its own after edits." If you want Claude to check the IDE's warnings, ask it to.

When selection context is active, the transcript shows a line like ⧉ Selected N lines from <file>. To keep a sensitive file such as .env out of that context, add a Read deny rule for its path. A matching rule blocks both the selected text and the open-file notice. See the permissions guide for how deny rules work.

How do you start and connect Claude Code from the IDE?

The simplest way is to open the IDE's integrated terminal and run claude. The docs say all integration features are then active.

You can also work from an external terminal, such as iTerm or Windows Terminal. Start Claude Code and run the /ide command:

claude
/ide

When it works, Claude Code confirms with a message like Connected to IntelliJ IDEA. Start Claude Code from the same directory as your IDE project root so both see the same files.

To switch where diffs appear, run /config and set Diff tool to auto (diffs in the IDE) or terminal (diffs stay in the terminal). That entry only shows up in /config while Claude Code is connected to the IDE, so connect first.

Which plugin settings are worth knowing?

The plugin settings live under Settings > Tools > Claude Code [Beta]. The general settings are:

• Claude command: a custom command to start Claude.
• Suppress notification for when Claude Command is not found: hides that warning.
• Enable using Option+Enter for multi-line prompts: macOS only. Turn it off if the Option key is captured unexpectedly. It needs a terminal restart.
• Enable automatic updates: checks for plugin updates and applies them on restart.

One common annoyance is the Esc key. By default, Esc in a JetBrains terminal can move focus back to the editor instead of interrupting Claude. The fix from the docs: go to Settings > Tools > Terminal and either uncheck "Move focus to the editor with Escape," or click "Configure terminal keybindings" and delete the "Switch focus to Editor" shortcut.

For JetBrains Remote Development, the docs warn that you "must install the plugin on the remote host via Settings → Plugin (Host)," not on your local client.

How do you fix Claude Code in JetBrains on WSL?

For WSL, set the Claude command to run inside your distribution. The docs give this example, where you replace Ubuntu with your distribution name:

wsl -d Ubuntu -- bash -lic "claude"

If /ide reports "No available IDEs detected" on WSL2, the docs say the cause is usually WSL2's NAT networking or Windows Firewall blocking traffic between WSL2 and the IDE on Windows. WSL1 is not affected. There are two documented fixes:

1. Allow WSL2 traffic through the firewall (recommended). Run hostname -I in WSL, take the first two parts of the address and add .0.0/16 to get your subnet. Then, in PowerShell as Administrator, create an inbound rule for that subnet:

New-NetFirewallRule -DisplayName "Allow WSL2 Internal Traffic" -Direction Inbound -Protocol TCP -Action Allow -RemoteAddress 172.21.0.0/16 -LocalAddress 172.21.0.0/16

2. Switch WSL2 to mirrored networking. This needs Windows 11 22H2 or later. Add networkingMode=mirrored under [wsl2] in .wslconfig in your Windows user folder, then run wsl --shutdown.

Restart the IDE and Claude Code after either change. The Claude Code on Windows guide covers the wider Windows setup.

What else can go wrong, and what are the security notes?

The troubleshooting section of the docs boils down to a short checklist:

• Plugin features missing: run Claude Code from the project root, check the plugin is enabled, and restart the IDE completely. The docs note you "may need to do this multiple times."
• IDE not detected: confirm the plugin is enabled, restart, and if you expected an automatic connection, check you launched claude from the IDE's own terminal.
• Command not found: run claude --version in a terminal, then set the full path in the Claude command setting.

On security, the docs flag one JetBrains-specific risk. In acceptEdits permission mode, Claude "may be able to modify IDE configuration files that can be automatically executed by your IDE," which could bypass the Bash permission prompts. They suggest using Manual mode for edits, using trusted prompts only, and staying aware of which files Claude can change.

Under the hood, the plugin runs a local MCP server named ide that the CLI connects to. It is hidden from /mcp, listens on 127.0.0.1 by default and uses a fresh auth token on each IDE start. Only one of its tools is visible to the model, mcp__ide__getDiagnostics, and the docs state "the JetBrains plugin does not expose a code-execution tool to the model." That matters if your team uses a PreToolUse hook to allowlist MCP tools.

Where do your prompts live when you switch IDEs?

The plugin shares your normal Claude Code setup, so settings, skills and CLAUDE.md files work the same in IntelliJ as in the terminal or VS Code. The part that tends to get lost is the prompts you type over and over: review requests, refactor instructions, test-writing templates.

Kipt keeps those prompts and your Agent Skills as Markdown in a private GitHub repo you own, with {{name}} or {{name:default}} blanks. You can install every skill into Claude Code with one npx skills add command, so the same library works whichever IDE you open.

If your team also uses VS Code, compare the Claude Code VS Code extension, which bundles its own CLI and adds a graphical chat panel.