# MCP Server Setup

> Connect Claude Desktop, Claude Code or Cursor to your Revnu agent with one config snippet.

Canonical: https://revnu.com/docs/mcp

`revnu mcp` is a stdio MCP server whose tools map one-to-one onto the API routes, so any MCP host can read your queue, approve or send back a card, and talk to your agent.

## Claude Desktop and Cursor

Add to `claude_desktop_config.json` or Cursor's MCP settings:

```json
{
  "mcpServers": {
    "revnu": {
      "command": "npx",
      "args": ["-y", "@revnu/cli", "mcp"],
      "env": { "REVNU_API_TOKEN": "rvp_..." }
    }
  }
}
```

## Claude Code

```bash
claude mcp add revnu -e REVNU_API_TOKEN=rvp_... -- npx -y @revnu/cli mcp
```

## Tools

`whoami`, `review_list`, `review_get`, `review_answer`, `review_send`, `review_request_changes`, `review_snooze`, `review_dismiss`, `agent_message` (posts, then waits for the reply), `agent_message_status`, `agent_thread`, `agent_status`, `results_get`, `workflows_list`, `workflow_runs`, `leads_list`, `lead_get`.

> **Careful.** Writes through an assistant are real writes: `review_answer` on a blog draft publishes it, `review_send` sends email. Mint a read-only token if your assistant should only look.

## Frequently asked questions

### Does the assistant see my whole account?

Only what the token's scopes allow. A token with `review:read` alone lets it list and read cards and nothing else.

### Can I run it against a staging API?

Yes. Add `REVNU_API_BASE` to the `env` block.