Keboola MCP Server is an open-source bridge that connects Keboola projects with modern AI tools, allowing users to expose data, transformations, SQL queries, and job triggers without requiring additional code.
From the registry: Connect your AI assistants to Keboola and expose your data, transformations, SQL queries, ...
$ brew install uvhttps://docs.astral.sh/uv/getting-started/installation/Please install the `mcp-keboola` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Get Keboola Storage Token** — Create a Storage API token in Keboola project settings (master token or custom token) (https://help.keboola.com/management/project/tokens/)
- **Install uv** — Install uv package manager for running uvx Run: `brew install uv` (https://docs.astral.sh/uv/getting-started/installation/)
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["keboola_mcp_server"]
- required environment variables:
- `KBC_STORAGE_API_URL`: Keboola region API URL (example: `https://connection.keboola.com`)
- `KBC_STORAGE_TOKEN`: Keboola storage API token (example: `<your-keboola-storage-token>`)
- optional environment variables:
- `KBC_WORKSPACE_SCHEMA`: Workspace schema (required only for custom tokens) (example: `<your-workspace-schema>`)
- `KBC_BRANCH_ID`: Dev branch ID (defaults to production) (example: `<branch-id>`)
Note: Remote hosted option also available at https://mcp.<REGION>.keboola.com/mcp with OAuth.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.KBC_STORAGE_TOKENrequiredAuthentication token for Keboola.KBC_STORAGE_API_URLrequiredAPI URL for your Keboola region.KBC_WORKSPACE_SCHEMArequiredIdentifies your workspace in Keboola for SQL queries.