Keboola
FreeKeboola 官方集成，在单一平台上构建数据工作流、集成和分析。
FreeFree tier
About Keboola
Keboola MCP Server is an open-source bridge that connects AI agents and MCP clients (such as Claude, Cursor, Windsurf, and VS Code) to the Keboola platform. It turns Keboola features—including storage access, SQL transformations, job triggers, workflow flows, and data apps—into callable tools that can be used via natural language. The server supports both local setup and a remote hosted version with OAuth authentication, enabling easy integration without glue code.
Key Features
Storage: query tables directly and manage table or bucket descriptions
Components: create, list, and inspect extractors, writers, data apps, and transformation configurations
SQL: create SQL transformations using natural language
Jobs: run components and transformations, retrieve job execution details
Flows: build and manage workflow pipelines with Conditional Flows and Orchestrator Flows
Data Apps: create, deploy, and manage Keboola Streamlit Data Apps that display queries over storage data
Metadata: search, read, and update project documentation and object metadata using natural language
Dev Branches: work safely in development branches scoped to the selected branch
Remote MCP Server: hosted solution with OAuth authentication for easy setup
Pros & Cons
Pros
- Open-source and free to use
- No glue code required to integrate AI agents with Keboola
- Supports multiple MCP clients: Cursor, Claude, Windsurf, VS Code, and more
- Remote server option provides quick setup with OAuth authentication
- Deep integration with Keboola platform features (storage, SQL, jobs, flows)
Cons
- Requires a Keboola project and active account to function
- Limited to the Keboola ecosystem and its data infrastructure
- May require familiarity with Keboola concepts (buckets, transformations, components)
Best For
Querying company data via natural language from an AI assistantAutomating data transformation jobs and pipelinesBuilding and managing complex workflow pipelines with conditional logicCreating interactive data apps for business usersManaging project metadata and documentation through conversational interfaces
FAQ
How do I connect to the remote MCP server?
Navigate to your Keboola Project Settings → MCP Server tab, copy the server URL (e.g., https://mcp.YOUR_REGION.keboola.com/mcp), paste it into your AI assistant's MCP settings, and authenticate with your Keboola account.
Which clients are supported?
Supported clients include Cursor, Claude Desktop, Claude Code, Windsurf, Make, and any MCP client supporting Streamable HTTP connections and OAuth.