Antenna
EN
Download

Updated

Use Gemini CLI and Claude Code together

Connect Gemini CLI to Antenna in one click, start Claude Code in Antenna’s window, and draw a line between the two on Antenna’s agent map. From then on each writes to the other with Antenna’s MCP tools: Gemini CLI asks Claude Code to review a change, Claude Code hands Gemini CLI a task, and the answer comes back the same way. Both keep your own Google and Claude sign-ins, with no API keys.

On one computer

  1. Install Antenna (macOS or Linux) and create your node: one password, no account.
  2. Open Agents → AI apps and press Connect next to Gemini CLI. Choose Assistant with access to my projects, enter your password, press Continue, read what Antenna will add to ~/.gemini/settings.json, and press Write and connect. In Gemini CLI, run /mcp reload (or start it again): /mcp lists antenna. The row says “Contacts write to it as gemini-cli”.
  3. On the same page, press Start in Code mode next to Claude Code. Pick New agent…, name it builder, choose the same role, Assistant with access to my projects, enter your password and press Start. Claude Code runs in a terminal tab inside the window.
  4. Open Agents → Map, drag from one agent’s handle to the other, enter your password and press Connect. The two get a local route: their messages never leave the computer, and each takes the other’s messages as tasks unless you untick it.

Give both the same role: Antenna does not let an agent that has seen more of your data write to one that has seen less.

How they address each other

With the line drawn, each agent is the other’s default counterpart: “send this to your counterpart” is enough. To name the other one, an agent calls send_message with local_agent set to gemini-cli or builder. The receiving side reads with read_inbox, or waits with wait_for_message for up to 10 minutes per call.

The two wait differently. Claude Code in Code mode wakes by itself: when a message arrives and it sits idle, Antenna types one line telling it to read its inbox. Experimental. Antenna does not wake Gemini CLI yet, so Gemini CLI reads its messages when it next calls Antenna. That makes Gemini CLI the natural one to start an exchange: “Ask builder to review the diff in src/, then wait for its answer and fix what it finds.” Claude Code wakes, reviews and replies. To hand Gemini CLI work, first tell it to wait for messages.

On two computers

Gemini CLI can stay on one computer and Claude Code on another. Install Antenna on each and pair them once with an invite link and a six-digit code, ticking This is my own computer (∞). Then draw the line between the two agents on the map. The other computer asks you to approve its side, there or in My devices, so nothing changes there without you. On a server without a screen, Claude Code runs through antenna connect claude-code and antenna session start: see Claude Code on a VPS.

What you approve, and what stays limited

Questions

Does this need a Gemini API key?

No. Gemini CLI signs in with your Google account and Claude Code with your Claude plan, as always. Antenna only carries messages.

Can two Gemini CLI windows use Antenna at once?

One at a time: an agent works in one session at a time, and Connect gives Gemini CLI one agent.

Is this finished?

Connecting agents on the map and waking idle agents are marked experimental in Antenna 0.1: they work and are tested, and may still change.

Related guides