Make finished videos from your Gemini CLI sessions.
Connect once, then ask Gemini CLI for a video — script, voiceover, visuals, and the final render land back in the conversation. Same credits as the app, no markup.
The consumer Gemini app doesn't take custom connectors yet — this works with Gemini CLI.
Connect Gemini CLI in minutes
- 01
Create an API key
In FacelessGenie, open Settings → API and create a key. It's shown once — copy it.
- 02
Add the server to settings.json
Add the snippet to ~/.gemini/settings.json (create the file if it doesn't exist).
- 03
Ask for your first video
In Gemini CLI, try "make a 60-second short about…" — it picks the format, creates the video, and shares the finished link when the render completes.
~/.gemini/settings.json
Swap in the key from Settings → API:
{
"mcpServers": {
"facelessgenie": {
"httpUrl": "https://www.facelessgenie.ai/api/mcp",
"headers": {
"Authorization": "Bearer fg_YOUR_API_KEY"
}
}
}
}Then ask for your first video — try “make a 60-second short about…”
A video team inside the chat
Create finished videos
Ask for a video on any topic — script, voiceover, visuals, captions, and the final render all happen for you.
Every format
Shorts, long-form documentaries, character shows, brainrot gameplay — the same catalog as the app.
Voices that don't sound like a robot
Pick narrators by name, or let the format's default carry it.
Use your own assets
Upload a character reference or your own music from the chat and use it in the next video.
Stay in control of spend
Preview the exact credit cost before creating, and check your balance any time.
You stay in charge
Access is tied to your own account, and you can cut it off anytime from Settings.
Questions, answered
Does the Gemini app support FacelessGenie?
Google's consumer Gemini app doesn't accept custom MCP connectors yet. Gemini CLI does — add FacelessGenie to ~/.gemini/settings.json with an API key and you're set.
Does creating videos from Gemini CLI cost extra?
No. You spend the same FacelessGenie credits as in the app — connecting never adds a markup. You can preview the exact cost of any video from the chat before creating it.
What can I do with FacelessGenie inside Gemini CLI?
Create complete faceless videos in any format, check render progress, preview credit costs, upload character references or music, browse your library, and check your balance — without leaving the conversation.
How long does a video take?
Renders typically take 5 to 20 minutes depending on length and format. The assistant can poll the status for you and share the finished mp4 link. Failed renders refund automatically.
Is the connection secure?
Yes. Access is tied to your own FacelessGenie account, and you can revoke it anytime from Settings.
Using a different tool? ChatGPT · Cursor · Claude Code · Windsurf · VS Code · Cline · Claude
Your next video is one message away.
Connect Gemini CLI and ask for your first video.