spiculedata/saiku

Open-source semantic layer: one cube for Excel (MDX/XMLA), dashboards, and AI agents (MCP). Mondrian + Apache Calcite.

What it solves

Saiku is an open-source semantic layer for analytics, allowing users to query complex data cubes without needing to write MDX or SQL. It bridges the gap between raw data warehouses and end-users (or AI agents) by providing a structured, typed interface for data exploration and visualization.

How it works

Saiku uses a combination of Mondrian (for OLAP cubes) and Apache Calcite (for SQL planning) to translate high-level queries into efficient database calls. It employs Apache Arrow for zero-copy data transfer to the browser. For AI integration, it exposes a typed REST API that allows LLM agents to discover schemas, translate natural language questions into validated MDX, and receive structured results. It also includes a Model Context Protocol (MCP) wrapper to integrate directly with AI tools like Claude Desktop or Cursor.

Who it’s for

  • Data Analysts: Who want a drag-and-drop browser interface for exploring cubes and building dashboards.
  • AI Developers: Who want to provide LLM agents with a reliable, typed way to query business data without the agents having to write raw SQL/MDX.
  • IT Operators: Who need a semantic layer with enterprise features like SCIM 2.0 user provisioning and OpenTelemetry observability.

Highlights

  • AI-Ready API: A dedicated REST surface for LLM agents with self-correcting validation errors.
  • MCP Support: Built-in Model Context Protocol wrapper for seamless AI agent integration.
  • Agent Skills & Spaces: No-code extension of AI capabilities via markdown-based workflows and persona-based scoping.
  • Semantic Model Generation: Ability to automatically build starting semantic models from a data warehouse.
  • Enterprise Tooling: Includes SCIM 2.0 provisioning, OIDC/SAML authentication, and a Google Sheets add-on.
  • Visual Analytics: Full dashboarding capabilities with cross-filtering, drill-downs, and anomaly/forecast overlays.

Related

  • Project
    cube-js/cube

    An open-source semantic layer that allows you to define metrics and business logic once in code and expose them via APIs to BI tools and AI agents.

  • Project
    okou-ai/okou

    Okou is an open-source agentic platform that automates cross-platform business workflows by allowing users to assign tasks to AI agents that connect to existing enterprise tools.

  • Project
    dsphper/lanhu-mcp

    Lanhu MCP Server is an open‑source bridge that implements the Model Context Protocol, letting AI coding assistants read Lan‑Hu design and requirement documents, extract visual and textual data with a vision‑capable LLM, and share results via a team knowledge board. It runs via HTTP or stdio, offers Docker and one‑click local install, and provides caching, incremental updates, and Feishu notifications.

  • Project
    agentscope-ai/agentscope

    A production-ready agent framework that provides an SDK and a FastAPI backend to build multi-tenant, multi-session agentic applications leveraging LLM reasoning and tool use.