> ## 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.

# Claude Code

> Connect ScrapeBadger to Claude Code (CLI) via MCP.

# Claude Code

Claude Code has native HTTP transport support — no bridge needed.

## Setup

Run this command in your terminal:

```bash theme={null}
claude mcp add --transport http scrapebadger \
  https://mcp.scrapebadger.com/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"
```

Or add to `.mcp.json` in your project root:

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

## Verify

Inside a Claude Code session, type `/mcp` to see connected servers. ScrapeBadger should show 92 tools.

## Scopes

| Scope           | Config Location              | Use Case                 |
| --------------- | ---------------------------- | ------------------------ |
| Local (default) | `~/.claude.json` per project | Project-specific         |
| Project         | `.mcp.json` in project root  | Shared with team via git |
| User            | `~/.claude.json` globally    | Available everywhere     |

Add `--scope user` to the `claude mcp add` command for global access.

## Examples

Once connected, ask Claude Code natural-language questions and it will pick the right tool:

> "Show me the top 5 results on Google for 'rust async runtime' and summarize them"

Claude Code calls `google_search`.

> "Find me a laptop under \$1500 on Google Shopping and give me the merchant URL for the cheapest one"

Claude Code chains `google_shopping_search` → `google_shopping_product_click`.