A DuckDB extension that calls Jev from SQL and returns answers as ENUM, numeric, or STRUCT types.
Real data. Real impact.
Emerging
Developers
git clone https://github.com/colliber/duckdb-jev.gitPer week
Excellent
One command unlocks capabilities that would take hours to build.
A DuckDB extension that calls Jev from SQL and returns answers as ENUM, numeric, or STRUCT types.
Asks Choice, Score, and Noul questions about row text and maps answers to SQL types defined by the questions.
Uses structured judgments while querying tables or Parquet data.
Requires an external Jev API configuration. SQL type constraints do not establish answer correctness or eliminate privacy risks.
Jev is TypeSafe AI's System One decision model: it returns typed scores and choices (not free text), so agents can route, filter and gate at low latency and cost.
Source: colliber/duckdb-jev — 23 stars, MIT. Listed by Awesome Jev.
MIT
4,600+ AI skills, agents & workflows. Install in 60 seconds. Part of the Torly.ai family.
© 2026 Torly.ai. All rights reserved.