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

# Claude Code

> Connect the RevyOps MCP server to Claude Code CLI.

## Prerequisites

* [Claude Code](https://claude.ai/claude-code) installed (`npm install -g @anthropic-ai/claude-code`)
* A RevyOps API key — get one from **Settings → API Keys**

***

## Setup

Add the RevyOps MCP server to your Claude Code config using the CLI:

```bash theme={null}
claude mcp add revyops \
  --transport http \
  --url https://mcp.revyops.com/mcp \
  --header "X-API-KEY: RO-your-api-key-here"
```

Or add it manually to your `~/.claude/settings.json`:

```json theme={null}
{
  "mcpServers": {
    "revyops": {
      "type": "http",
      "url": "https://mcp.revyops.com/mcp",
      "headers": {
        "X-API-KEY": "RO-your-api-key-here"
      }
    }
  }
}
```

***

## Verify

Run Claude Code and check that the server is connected:

```bash theme={null}
claude
```

Then in the session:

```
> List my RevyOps clients
```

Claude should respond with your available workspace(s).

***

## Multiple Workspaces

Pass a comma-separated list of API keys to access multiple workspaces in one session:

```bash theme={null}
claude mcp add revyops \
  --transport http \
  --url https://mcp.revyops.com/mcp \
  --header "X-API-KEY: RO-key-a,RO-key-b"
```

Use `list_clients` to see all resolved workspaces, then specify by name in your prompts:

```
> Show me campaigns for Client B from last week
```

***

## Project-Scoped Config

To scope the MCP server to a specific project (instead of globally), add a `.claude/settings.json` file in your project root:

```json theme={null}
{
  "mcpServers": {
    "revyops": {
      "type": "http",
      "url": "https://mcp.revyops.com/mcp",
      "headers": {
        "X-API-KEY": "RO-your-api-key-here"
      }
    }
  }
}
```

This keeps the connection isolated to that project's Claude Code sessions.