Skip to main content
Gemini CLI is Google’s command-line AI assistant. You can connect it to Kavio’s MCP server to manage your social media scheduling with natural language.

Prerequisites

Setup

1

Open the settings file

Edit your Gemini CLI settings at ~/.gemini/settings.json:
With oauth.enabled, Gemini CLI discovers Kavio’s OAuth metadata from the URL and registers dynamically. For project-level configuration, create .gemini/settings.json in your project root instead.
2

Restart Gemini CLI and sign in

Close and reopen Gemini CLI. The first Kavio call prompts you to sign in — Gemini CLI prints the authorize URL or opens your browser. Approve and return to the terminal.
3

Verify the connection

Ask Gemini to interact with your Kavio account:
Gemini CLI uses "httpUrl" for HTTP-streaming servers (not "url"). The "url" field is used for SSE transport only.
For headless environments, use a Personal Access Token instead of OAuth:
Create the key from Settings > API Keys on the Kavio dashboard.

Example prompts