Orionbelt Semantic Layer

What is Orionbelt Semantic Layer?

OrionBelt Semantic Layer is an API-first engine that transforms declarative YAML semantic models into optimized SQL for Postgres, Snowflake, ClickHouse, Dremio, and Databricks. It provides a unified abstraction over data warehouses so analysts and applications can query using business concepts (dimensions, measures, metrics) instead of raw SQL.

How to use Orionbelt Semantic Layer?

Clone the repository, install dependencies with uv sync, then start the REST API with uv run orionbelt-api (available at http://127.0.0.1:8000) or the MCP server with uv run orionbelt-mcp. For Claude Desktop, add the server to claude_desktop_config.json. Optionally install the Gradio UI with uv sync --extra ui and access it at /ui.

Key features of Orionbelt Semantic Layer

• 5 SQL dialects: Postgres, Snowflake, ClickHouse, Dremio, Databricks
• AST‑based SQL generation (no string concatenation)
• YAML semantic models with dimensions, measures, metrics, and joins
• Automatic join path resolution with Composite Fact Layer support
• Vendor‑specific SQL validation via sqlglot (non‑blocking)
• Precise error reporting with YAML source positions and join graph analysis
• TTL‑scoped session management via REST API and MCP
• ER diagram generation (Mermaid) via API and Gradio UI
• 9 MCP tools + 3 prompts for AI‑assisted model development
• Gradio UI for interactive model editing and SQL compilation

Use cases of Orionbelt Semantic Layer

• Compile business‑friendly queries (dimensions/measures) into dialect‑specific SQL
• Integrate with AI assistants (Claude Desktop, Cursor) for semantic model authoring
• Provide a unified semantic layer across multiple SQL databases
• Validate and debug YAML model definitions with precise error messages
• Generate ER diagrams from semantic models for documentation

FAQ from Orionbelt Semantic Layer

Which SQL dialects are supported?

Postgres, Snowflake, ClickHouse, Dremio, and Databricks SQL, each with dialect‑specific optimizations.

How do I run the MCP server?

Run uv run orionbelt-mcp for stdio mode (default, used with Claude Desktop) or set MCP_TRANSPORT=http for HTTP transport.

What tools and prompts does the MCP server expose?

9 tools: create_session, close_session, list_sessions, load_model, validate_model, describe_model, compile_query, list_models, list_dialects. 3 prompts: write_obml_model, write_query, debug_validation.

What are the prerequisites for installation?

Python 3.12+ and the uv package manager.

Is there a user interface besides the CLI and API?

Yes, a Gradio UI is available. Install with uv sync --extra ui and access it at /ui when the REST API server is running.