JetBrains IDEs Setup

Overview

JetBrains IDEs — IntelliJ IDEA, PyCharm, WebStorm, GoLand, and the rest of the family — support MCP through AI Assistant. Connect Tempreon once and AI Assistant loads your Core Imprint, searches your Knowledge Vault, and remembers what it learns, across every tool you use. The same server also works with Junie, JetBrains' autonomous coding agent.

How Tempreon connects here. AI Assistant's MCP client does not run a browser OAuth flow — JetBrains documents only a pasted bearer token or a local process, and an open JetBrains tracker item confirms OAuth2 is not supported for MCP server connections. Tempreon authorizes over OAuth, so on AI Assistant it connects through the mcp-remote proxy, which handles the browser approval outside the IDE. In Junie, which does document browser authorization, the plain url form works directly. Both are below.

Quick path: Settings | Tools | AI Assistant | Model Context Protocol (MCP) → Add → paste the mcp-remote config → approve in the browser tab it opens.

Prerequisites

• A JetBrains IDE with the AI Assistant plugin enabled. A remote server configured by url needs 2025.3 or later; 2025.2 supports stdio only. The mcp-remote config below is stdio, so it works on either
• Node.js available, so npx can run the mcp-remote proxy
• A JetBrains AI tier — there's an AI Free tier with a small cloud quota, plus paid AI Pro, Ultimate and Enterprise tiers; AI Pro also comes with the All Products Pack or dotUltimate, though it can't be activated when the IDE license comes through a License Server. AI Free is not available in Community Edition IntelliJ IDEA or PyCharm, in PyCharm Unified used for free, or in Android Studio, so those need a paid tier
• A Tempreon account with at least one available Bridge slot
• The Tempreon server URL (also shown in your Bridges dashboard)

Step-by-Step Setup

1. Open the MCP settings

1. Open Settings
2. Go to Tools | AI Assistant | Model Context Protocol (MCP)
3. Click Add

The Level setting decides whether Tempreon is available globally (all projects) or only in the current project. If you already use Tempreon in Claude Desktop, the Import from Claude button can pull that config in directly.

2. Paste the configuration

Choose the JSON option and paste. mcp-remote speaks stdio to the IDE and performs Tempreon's OAuth handshake in your browser:

{
  "mcpServers": {
    "tempreon": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://api.tempreon.com/functions/v1/tempreon-mcp/mcp"]
    }
  }
}

Click OK, then Apply.

3. Authorize the connection

A browser tab opens to Tempreon's consent screen — sign in to your Tempreon account and approve, and the proxy stores the token. Tempreon supports dynamic client registration, so there is no client ID or secret to enter.

If the tab doesn't open, run the same command once in a terminal to complete the handshake interactively:

npx -y mcp-remote https://api.tempreon.com/functions/v1/tempreon-mcp/mcp

4. Confirm the tools are active

The server should show a connected status in the MCP settings list. In an AI Assistant chat, ask "What Tempreon tools do you have access to?" If it answers with a list, the Bridge is live.

Junie

Junie is the JetBrains agent that documents browser authorization, so it can take Tempreon's URL directly — no proxy needed.

Give Junie its own config — .junie/mcp/mcp.json in the project root, or ~/.junie/mcp/mcp.json for user-level:

{
  "mcpServers": {
    "tempreon": {
      "url": "https://api.tempreon.com/functions/v1/tempreon-mcp/mcp"
    }
  }
}

Tempreon then appears in Junie's server list with an Authorization required status. Select it, choose Authorize, and sign in on the Tempreon page that opens.

Or reuse the AI Assistant server. Configure Tempreon under Settings | Tools | AI Assistant | Model Context Protocol (MCP) as above, then go to Settings | Tools | AI Assistant | Agents, enable Pass custom MCP servers, and restart the IDE. Note this passes through the mcp-remote entry, proxy and all — if you want Junie to authorize Tempreon directly, use its own config file instead.

Junie's documented remote-server entry is a url plus optional headers — there is no required type key. The "type": "streamable-http" snippet that appears elsewhere in JetBrains' docs points Junie at the IDE's own MCP server for debug mode, which is the opposite direction and needs IntelliJ IDEA 2026.1.1 or newer.

Troubleshooting

Connection fails immediately. The URL must end in /mcp with no trailing slash: https://api.tempreon.com/functions/v1/tempreon-mcp/mcp.

Don't see the MCP settings page? Confirm the AI Assistant plugin is installed and enabled, and that you have an active JetBrains AI tier. Note AI Free is not offered in Community Edition IntelliJ IDEA or PyCharm, in PyCharm Unified used for free, or in Android Studio, so on those a paid tier is required.

You pasted Tempreon's url directly and nothing authorizes. Expected on AI Assistant — it has no browser OAuth flow, so a bare url entry for an OAuth-protected server has no way to complete. Use the mcp-remote config from step 2, or connect Tempreon through Junie.

npx not found, or the proxy never starts. mcp-remote runs through Node.js — install Node, then reopen the IDE so it picks up the updated PATH.

The url form is rejected in Junie's config. Junie takes url on any current build; if the IDE itself is on 2025.2, AI Assistant's own list supports stdio only, which is why step 2 uses the proxy shape there.

Also connect: VS Code · Zed · Windsurf

Try saying...

• "Start my session." — triggers session_start to load your Core Imprint
• "Search my knowledge base for our testing conventions." — exercises knowledge retrieval end-to-end
• "Remember that we target Java 21 across all services." — stores a decision for every future session