Frequently Asked Questions (FAQ)

Q: What is the Keboola MCP Server? A: The Keboola MCP Server is an open-source bridge that connects your Keboola project to modern AI tools, such as Claude and Cursor. It allows AI agents to access and interact with your Keboola data, transformations, and jobs.

Q: What is MCP? A: MCP stands for Model Context Protocol. It is an open protocol that standardizes how applications provide context to Large Language Models (LLMs).

Q: What are the benefits of using the Keboola MCP Server? A: The Keboola MCP Server offers several benefits, including seamless integration with AI tools, simplified data access for AI agents, automated data pipeline management, and improved collaboration between AI models and your data infrastructure.

Q: What MCP clients are supported? A: The Keboola MCP Server supports various MCP clients, including Claude (Desktop & Web), Cursor, Windsurf, Zed, Replit, Codeium, Sourcegraph, and custom MCP clients.

Q: What Keboola features can I access through the MCP Server? A: The MCP Server provides access to Keboola storage (tables, buckets), components (extractors, writers, data apps, transformations), SQL transformations, jobs, and metadata.

Q: How do I configure the Keboola MCP Server? A: You can configure the MCP Server using one of the available methods: Integrated Mode (recommended), Local Development Mode, Manual CLI Mode, or Docker. Configuration typically involves providing your Keboola Storage API token, workspace schema, and other relevant settings.

Q: Do I need to start the MCP Server manually? A: It depends on your setup. If you are using Claude or Cursor, the MCP Server can be started automatically by the client. If you are developing locally or using Docker, you may need to start the server manually.

Q: What if I am using BigQuery? A: If your Keboola project uses BigQuery, you will need to set the GOOGLE_APPLICATION_CREDENTIALS environment variable to point to your BigQuery credentials file.

Q: Where can I get support for the Keboola MCP Server? A: The primary way to get help, report bugs, or request features is by opening an issue on the GitHub repository.