Vibe Coding Discoverorionbelt-semantic-layer
View on GitHubSource-available semantic and context layer, rule engine, and semantic sidecar for agentic AI and governed analytics. Compiles declarative YAML models into optimized SQL, KPIs, and semantic context across 8 SQL dialects.
OrionBelt compiles YAML semantic models, metrics, joins, and business rules into SQL across eight database dialects. It exposes governed data context and query access to AI agents through MCP, as well as BI tools through REST, PostgreSQL wire, and Arrow Flight SQL.
Use Cases
Let AI agents query governed business metrics through MCPGenerate dialect-specific SQL from declarative YAML modelsPrevent fan-trap errors in multi-fact analytics queriesExpose business rules and metric definitions to agentsServe semantic models to BI tools over PostgreSQL wire or Arrow Flight SQLCompile analytics queries for BigQuery, ClickHouse, Databricks, Dremio, DuckDB, MySQL, PostgreSQL, and Snowflake
Built With
- Language
- Python
- Frameworks
- FastAPI · Pydantic · SQLGlot · NetworkX · RDFLib · DuckDB
Tags
semantic layer · context layer · MCP server · agentic AI · text-to-SQL · analytics · metrics layer · business rules · SQL generation · YAML models · data governance · multi-dialect SQL