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

# Gemini CLI

> Connect ScrapeBadger to Google Gemini CLI via MCP.

# Gemini CLI

## Setup

Edit `~/.gemini/settings.json` (or `.gemini/settings.json` in your project):

```json theme={null}
{
  "mcpServers": {
    "scrapebadger": {
      "httpUrl": "https://mcp.scrapebadger.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}
```

<Warning>
  Gemini CLI uses `httpUrl` — not `url`. This is different from Cursor and other tools.
</Warning>

## Verify

Run `gemini` and ask it to use a ScrapeBadger tool. The tools will be discovered automatically.

<Note>
  The Gemini web app (gemini.google.com) does **not** support custom MCP servers. MCP is available through the Gemini CLI only.
</Note>

## Examples

ScrapeBadger gives Gemini CLI 92 tools across Twitter, Google, Vinted, and Web Scraping. Try:

```bash theme={null}
gemini "Search Google for 'best react state management 2026' and list the top 5 results"
```

```bash theme={null}
gemini "Find a Razer gaming headset on Google Shopping and resolve the merchant URL"
```

Gemini will discover and call the appropriate `google_search` / `google_shopping_*` tools.