The Debitura MCP server lets you connect an MCP-compatible AI assistant, such as Claude, Cursor, or VS Code Copilot, to your Debitura account, so you can check cases, preview pricing, and submit new collection cases in plain language. This guide shows you how to connect one.

You need:

The MCP server uses the same per-account API key as the Debitura Customer API. The key identifies which account the assistant acts for, so treat it like a password and keep it secret.

Create your key on the Debitura API key page. For the full steps, see Clients: How to generate an API key.

In your AI client, add a new custom MCP connector that points at the Debitura endpoint and sends your API key in the XApiKey header:

If your client only supports a bearer token field instead of a custom header, you can send the same key as Authorization: Bearer followed by your key.

The exact steps differ per client. In Claude (web or desktop), open Settings, then Connectors, then Add custom connector, paste the endpoint URL, and add your API key. For ready-made setup snippets for Claude Code, Cursor, and VS Code Copilot, follow the authoritative instructions in the Debitura MCP repository.

Once connected, ask your assistant: "Ping Debitura". It should reply "Connected as" followed by your company name. If it does, the connection is working.

Your AI assistant can now work with your Debitura account. You can ask it to list your cases, check a case status, read messages with a collection partner, preview pricing, or submit a new case. Submitting a case always asks you to confirm first. To understand everything you can do and the safeguards involved, see Clients: Using Debitura from an AI assistant (MCP server).

If "Ping Debitura" does not return your company name, check that the endpoint is exactly https://mcp.debitura.com/mcp, the header name is XApiKey, and your API key is active on the Debitura API key page. If it still does not connect, contact support and include the name of the AI client you are using.