Skip to main content Skip to complementary content

Connecting to the Qlik MCP server

This section outlines how to connect using some of the most common LLM clients:

  • OpenAI ChatGPT

  • Anthropic Claude Desktop

Qlik Cloud Government note

The predefined OpenAI Chat GPT and Anthropic Claude Desktop clients below are configured for commercial implementation and not government implementations. Qlik Cloud Government and Qlik Cloud Government - DoD users must create their own OAuth clients. Please refer to Claude for Government and ChatGPT for Government documentation, and Connecting to other LLM clients below.

Qlik Cloud Government noteQlik Cloud Government and Qlik Cloud Government - DoD: To use Qlik MCP, you must enable cross-region data processing. MCP may be available before other AI features are released. Fore more information, see Enabling cross-region data processing.

Connecting to ChatGPT

Connect to the Qlik MCP server by creating an app in ChatGPT.

Information note

You must be on a ChatGPT Team, Pro, or Plus plan to connect to the Qlik MCP.

  1. In ChatGPT, in Settings > Apps > Advanced Settings, turn on Enable Developer Mode.

  2. Click Create App.

  3. After Name, enter a name for your connection.

  4. After MCP Server URL, enter <your tenant URL>/api/ai/mcp.

  5. Under Authentication, select OAuth.

  6. For the Client ID, enter 0bea8559a554c4d0de177952d8e57f62.

  7. For the Client Secret, leave the field empty.

  8. Click Create.

  9. Sign into your Qlik Cloud tenant.

  10. Click Approve.

    Warning noteBy clicking Approve, you will enable the Qlik Server to share your Qlik Cloud data ("Content") with a third-party system (like an AI tool). That system is not managed by Qlik, and your Content will be located, stored and processed under the third party’s control.

Connecting to Claude

Connect to Qlik MCP server by creating a custom connector in Claude.

video thumbnail

  1. In Claude, open Settings.

  2. In Connectors, click Add custom connector.

  3. Enter a name for the custom connector, such as Qlik MCP.

  4. For the Remote MCP server URL, enter <your tenant URL>/api/ai/mcp.

  5. Under Advanced settings, in OAuth Client ID, enter 76d3f46e87655a50424bec7e0f0bb1e2.

  6. For the OAuth Client Secret, leave the field empty.

  7. Click Add.

  8. Click Connect on your custom connector.

  9. Sign into your Qlik Cloud tenant.

  10. Click Approve.

    Warning noteBy clicking Approve, you will enable the Qlik Server to share your Qlik Cloud data ("Content") with a third-party system (like an AI tool). That system is not managed by Qlik, and your Content will be located, stored and processed under the third party’s control.

Connecting to other LLM clients

You can connect other LLM clients to the Qlik MCP server. A tenant administrator must create an OAuth client for the LLM client. Once the OAuth client is ready, you can add the Qlik MCP server as a remote connection. For information on creating the OAuth client, see Creating OAuth clients for LLM clients.

The section offers general guidelines for adding the Qlik MCP server to an LLM client. Specifics will vary depending on your LLM client.

  1. In your LLM client, add a new remote connection.

  2. For the URL, enter <your tenant URL>/api/ai/mcp.

  3. For the Client ID, enter the Client ID of the OAuth client.

  4. Create the connection.

  5. Sign into your Qlik Cloud tenant.

  6. Click Approve.

    Warning noteBy clicking Approve, you will enable the Qlik Server to share your Qlik Cloud data ("Content") with a third-party system (like an AI tool). That system is not managed by Qlik, and your Content will be located, stored and processed under the third party’s control.

Limitations

Qlik MCP server has the following limitations:

  • Qlik MCP sessions are subject to OAuth timeout settings governed by Qlik. If your session is disconnected, access the LLM client's settings to establish a new session.

Did this page help you?

If you find any issues with this page or its content – a typo, a missing step, or a technical error – please let us know!