Use Gemini CLI with Pluely

Connect Gemini CLI — Google's open-source AI CLI with a generous free tier — and the Pluely overlay answers through your own Google account, locally.

1. Install and sign in

Install Gemini CLI (for example npm install -g @google/gemini-cli) and run gemini once in your terminal to sign in with Google. If it answers in your terminal, Pluely can use it.

2. Connect it in Pluely

1. Open the dashboard → Dev space → Connect your CLI.
2. Pluely detects Gemini CLI automatically. Click Connect.
3. It appears at the top of the overlay's model picker as Your CLI.

Pluely handles Gemini CLI's workspace trust requirement for you — no "not running in a trusted directory" errors, no flags to set.

3. Pick a model

The Model panel sits right under the connected row:

• Default (CLI decides) — your Gemini CLI's configured model. Always works.
• Gemini 3 Pro (preview), Gemini 2.5 Pro, Gemini 2.5 Flash — one click each.
• Custom model id — any exact model id your account can access.

Images work

Gemini is vision-capable through Pluely: attached screenshots and images ride along with your question automatically.

Thinking models work

Gemini's thinking output is filtered automatically — you see only the final answer.

Troubleshooting

• "Your CLI couldn't answer" naming a model — your account may not have access to that model (previews roll out gradually); switch to Default or 2.5 Pro.
• Sign-in expired — run gemini in a terminal to re-authenticate.

Related

• Connect a local CLI (all options)
• Choosing your AI