pgenie.io

What It Is

pGenie is a SQL-first PostgreSQL development tool designed to preserve the readability and reviewability of plain SQL while adding type safety, query validation, and data access layer generation. It reads PostgreSQL migrations from migrations/ and queries from queries/. After running pgn generate, it starts a temporary database or connects to an existing PostgreSQL instance, applies migrations, analyzes queries, and generates a type-safe client API.

Core Capabilities and Ecosystem

Its core model is neither an ORM nor a query builder; instead, it treats the database schema and SQL queries as the source of truth. pGenie can validate SQL against the real schema, infer parameter types, result column types, and nullability, detect schema/query mismatches caused by migrations, and fail PRs in CI. The site also highlights index management capabilities: generating migrations to add missing indexes and remove redundant ones. It currently explicitly supports generating Haskell, Rust, and Java SDKs. Custom generators use Dhall and can be extended to other languages or frameworks. The documentation covers installation, project initialization, migrations, queries, CI/CD, AI, Flyway/Liquibase/Sqitch integrations, and comparisons with sqlc and jOOQ, with a fairly complete structure.

Pricing and Deployment

The crawled content does not disclose pricing, paid plans, payment methods, or commercial support. Its deployment model appears closer to a local/CI tool: by default it can start a temporary PostgreSQL instance via Docker, and it also supports using an existing instance with --database-url. No cloud-hosted service information was found.

Pros and Cons

The advantages are that SQL stays native, static code is generated, there is no runtime abstraction overhead, and many database errors can be shifted earlier into the generation stage. This is especially useful for reviewing AI-generated SQL. The limitations are also clear: it only supports PostgreSQL, so it is not suitable for multi-database projects; it only handles static, prewritten parameterized SQL, making it less suitable for complex dynamic reports or user-driven queries; and the extra generation step may add friction for early-stage rapid prototyping. Its maturity is also not yet on par with traditional ORM ecosystems.

Who It’s For and Access from China

It is a good fit for backend teams that are comfortable with SQL, care about the safety of production changes, and need data access layers across multiple languages. The crawled text does not provide information about access from China, so this is currently unknown. If access to GitHub, documentation, or images is affected by network conditions, alternatives such as sqlc, jOOQ, ORMs, or query builders may be worth evaluating.