Model Context Protocol (MCP) Overview & Setup

The Model Context Protocol (MCP) allows you to securely connect AI assistants (like Claude Desktop) to your Revelator data. By configuring these servers, your Large Language Models (LLMs) can execute specific tools to fetch, analyze, and manage your catalog, analytics, and royalty data in real-time using natural language.

Available Servers

We have logically grouped our tools into five distinct MCP servers based on business domains. Choose a server below to view its connection address and supported tools:

1. Catalog MCP Server: Query and manage metadata for releases, tracks, and artists.
2. Supply Chain MCP Server: Manage distribution settings, pricing, and territories.
3. Analytics MCP Server: Pull performance data, Top Movers, and revenue metrics.
4. Royalties MCP Server: Manage accounting, royalty runs, and payee statements.
5. Blockchain MCP Server: Interact with Web3 features, royalty tokens, and smart wallets.

---

Remote Setup Instructions

To connect your AI agent to Revelator’s data, you must add the appropriate MCP Server to your client’s configuration file using a Streamable HTTP transport.

Step 1: Add the Server Configuration

Point your AI agent to the specific MCP Server endpoint for the domain you want to access (Catalog, Analytics, etc.).

For example, to enable the Catalog tools, add the following to your client’s JSON configuration file:

{
  "mcpServers": {
    "revelator-catalog": {
      "type": "http",
      "url": "https://mcp.revelator.com/catalog"
    }
  }
}

Note: You can configure multiple Revelator servers simultaneously by adding additional entries to the mcpServers object using their respective URLs.

Step 2: Global Setup & Authentication

Because your AI assistant is executing actions on your behalf, it must be properly authenticated.

Use your existing Revelator Partner API credentials (partnerUserId and partnerApiKey) when your LLM prompts you to do so. This is usually only required initially when starting a new chat/session.

---

Supported Clients

You can connect Revelator MCP Servers to any AI assistant, IDE, or agent framework that supports the Model Context Protocol. Below is a list of popular supported clients:

Client	Developer	Notes
Claude Desktop	Anthropic	Full native MCP support. Configure via claude_desktop_config.json.
Claude Cloud (Web)	Anthropic	Requires an Enterprise/Team plan for MCP integrations.
ChatGPT Desktop	OpenAI	Supports MCP tools via local app configuration.
ChatGPT Cloud (Web)	OpenAI	Supported via custom GPT actions or enterprise integrations.
Cloudflare AI Playground	Cloudflare	Web-based environment perfect for testing MCP connections.
Visual Studio Code	Microsoft	Supported via extensions (e.g., Cline, Roo Code). Configure via cline_mcp_settings.json.
Cursor	Cursor	Add MCP servers directly in the IDE settings to code alongside your catalog data.
Claude Code	Anthropic	CLI-based agent for terminal usage.
Gemini CLI	Google	CLI-based agent. Configure via ~/.gemini/settings.json.
Kiro CLI	Kiro	CLI-based agent. Configure via ~/.kiro/settings/mcp.json.