Gemini CLI Setup

Overview

Gemini CLI is Google's open-source terminal agent for Gemini models. It supports remote MCP servers with OAuth, so it can connect to Tempreon with a single command or a small settings.json entry.

Connecting Tempreon gives your terminal sessions persistent intelligence: your AI loads your Core Imprint, searches your Knowledge Vault for context, and remembers what it learns — and that memory is shared with every other tool you've connected.

Quick path: gemini mcp add --transport http tempreon <URL> -s user → /mcp auth tempreon → approve in the browser.

Availability (as of July 2026): On June 18, 2026, Gemini CLI stopped serving requests for consumer Google accounts — free, Google AI Pro, and Ultra. It continues to work with Gemini Code Assist Standard or Enterprise licenses and paid Gemini API keys. If you're on a consumer account, use Google Antigravity instead — it's Google's consumer replacement and has its own guide.

Prerequisites

• Gemini CLI installed, running against an account it still serves: a Gemini Code Assist Standard or Enterprise license, or a paid Gemini API key (see the availability note above)
• A Tempreon account with at least one available Bridge slot
• A browser for the authorization step

Step-by-Step Setup

1. Add the server

The fastest path is one command:

gemini mcp add --transport http tempreon https://api.tempreon.com/functions/v1/tempreon-mcp/mcp -s user

-s user writes the entry to your user-level ~/.gemini/settings.json, so it's available in every project.

Prefer editing the file directly? Add this to ~/.gemini/settings.json (or a project's .gemini/settings.json for one workspace):

{
  "mcpServers": {
    "tempreon": {
      "httpUrl": "https://api.tempreon.com/functions/v1/tempreon-mcp/mcp",
      "oauth": { "enabled": true }
    }
  }
}

Use httpUrl, not url. In Gemini CLI, url means an SSE server — httpUrl is the Streamable HTTP transport Tempreon uses. The "oauth": {"enabled": true} block is optional, but it makes the auth expectation explicit.

The URL is also shown in your Bridges dashboard.

2. Authorize the connection

Start Gemini CLI and run:

/mcp auth tempreon

A browser tab opens to Tempreon's consent screen — sign in to your Tempreon account and approve. Tokens are stored in ~/.gemini/mcp-oauth-tokens.json.

3. Verify it's working

Run /mcp to check server status, or ask "What Tempreon tools do you have access to?" — if it answers with a list, the Bridge is live.

Gemini Code Assist

The Gemini Code Assist VS Code extension's agent mode reads the same ~/.gemini/settings.json (Standard and Enterprise editions), so the config above carries over with no extra steps. If the OAuth flow doesn't complete inside the IDE, run the authorization once in Gemini CLI (/mcp auth tempreon) — settings and tokens are shared, so the IDE picks up the authenticated server.

Troubleshooting

Requests aren't served on a personal Google account. That's the June 18, 2026 consumer retirement, not a Tempreon problem — some third-party guides still describe the free tier, but that information is out of date. Use Google Antigravity instead.

Connection fails immediately. Confirm the URL ends in /mcp with no trailing slash: https://api.tempreon.com/functions/v1/tempreon-mcp/mcp.

The server connects but tools fail or auth never triggers. Check the config field: it must be httpUrl, not url (which Gemini CLI treats as SSE). Then run /mcp auth tempreon.

"Unauthorized" or stale-token errors. Run /mcp auth tempreon again to re-authorize. Tokens live in ~/.gemini/mcp-oauth-tokens.json.

OAuth won't complete inside VS Code. Run the auth once in Gemini CLI — the shared settings mean the extension inherits the authenticated connection.

Also connect: Google Antigravity · Gemini App · Any MCP Client

Try saying...

• "Start my session." — triggers session_start to load your Core Imprint
• "Search my knowledge base for [topic] before we plan this change." — pulls your Knowledge Vault into the session
• "Remember that [decision] — we chose it because [reason]." — stores the insight for every future session, in every connected tool