Mnemiq
Open-source text-to-SQL you can tune and measure on your own database
Emerged finds new products worth a look, daily — what’s new today
You get Cube-style metric definitions and row-level security compiled into SQL, without giving a third party credentials to your warehouse.
0sql takes a YAML semantic model you deploy from its CLI, then turns JSON query specs of measures, dimensions, filters and cohorts into one SQL statement you run yourself. It never connects to your data, keeps no logs of specs, and bakes row-level security into the output. The same spec always yields the same SQL, across 13 dialects including Postgres, Snowflake, BigQuery and Databricks. It also exposes discovery endpoints and typed errors for agents. Free while in development; built by Strata Business Intelligence.
More like this
Musql
Embedded Go SQL with SQLite's dialect, columnar storage and a query JIT
Graphene
Semantic layer and SQL-in-Markdown dashboards your coding agent can edit in a PR
SchemaGate
Shows a text-to-SQL model only the tables the caller is allowed to see
Omnesis
A self-hosted index of your email, chats, health and files for any AI agent
xarray-sql
Run SQL over Xarray datasets with DuckDB, DataFusion or Polars