> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentbees.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Gemini CLI

> Google's coding CLI — runs on a Google API key.

**Gemini CLI** (`gemini`) is Google's agentic coding CLI. It authenticates with a
**Google API key**.

## Set it up

<Steps>
  <Step title="Open Admin → Agent LLM">Scroll to the per-agent keys section.</Step>

  <Step title="Add the Google key (Gemini)">
    Paste your Google API key. It's **write-only and encrypted at rest**, and only needed if you
    enable the Gemini agent.
  </Step>

  <Step title="(Optional) allow specific models">
    Set a **Gemini model allowlist** (e.g. `gemini-2.5-pro, gemini-2.5-flash`). With two or more,
    the New Task form shows a Model dropdown for Gemini tasks.
  </Step>
</Steps>

## Using it in a task

Pick **Gemini CLI** as the agent in the New Task form, and choose the model per
task if an allowlist is set.

## Through the internal gateway

When routed through the [internal gateway](/agents/internal-gateway), Gemini uses
the gateway **origin** (like Claude) — derived automatically from your single
gateway endpoint. A per-agent key is only needed if the gateway issues a
different key per model family.

<Tip>
  Gemini also appears as a **resume option** for a paused task — you can hand a stalled task from
  one CLI to Gemini in place. See [Coordination](/workflows/coordination#cross-agent-handoff).
</Tip>