Open SourceFrameworkFree

WrenAI

Canner's context layer for querying business data in natural language: you describe entities, relationships, and metrics once in a semantic model (their MDL), and any agent queries through that governed layer across 20-plus sources (PostgreSQL, BigQuery, Snowflake, Databricks, ClickHouse, DuckDB, and more), with a Rust engine on Apache DataFusion doing the translation. Apache-2.0 for the code (docs CC-BY-4.0, AGPL-3.0 held in reserve for future modules). Restructured May 2026: the old turnkey GenBI web app now lives on the legacy/v1 branch, and main is the context layer and SDK.

Homepage ↗Repo ↗

1 workflow use WrenAI

AgentsFreeMachine-verified

Validate a WrenAI semantic model's references before an agent queries through it

Before letting an agent query through WrenAI's governed semantic layer (MDL), validate that every relationship and metric resolves to a model and column that actually exist, so a stale definition fails a check instead of quietly returning wrong-but-plausible numbers.

WrenAI· intermediate