# GizmoSQL

> GizmoSQL is a high-performance SQL server for the cloud, built on Apache Arrow Flight SQL with DuckDB and SQLite backends. It provides authenticated, TLS-secured, remote SQL execution with JDBC, ADBC, Python, and CLI clients.

Each page below is available as raw Markdown at the listed URL (the rendered
HTML version lives at the same path without the `.md` extension).

## Getting Started

- [Documentation home](https://docs.gizmosql.com/index.md): overview, installation, deployment, configuration, and client connection examples
- [Quick Start](https://docs.gizmosql.com/quickstart.md): get a GizmoSQL server running in minutes
- [Editions](https://docs.gizmosql.com/editions.md): Core vs Enterprise feature comparison
- [LTS Channel](https://docs.gizmosql.com/lts_channel.md): long-term support release channel
- [Edge Channel](https://docs.gizmosql.com/edge_channel.md): experimental channel on the next DuckDB major (pre-release; not for production)
- [Security Guide](https://docs.gizmosql.com/security.md): TLS, authentication, and hardening guidance
- [Python ADBC Driver](https://docs.gizmosql.com/python_adbc.md): connect from Python via ADBC
- [Python Library](https://docs.gizmosql.com/python_library.md): run a GizmoSQL server from Python
- [Client Shell](https://docs.gizmosql.com/client.md): the gizmosql_client command-line shell
- [Token Authentication](https://docs.gizmosql.com/token_authentication.md): JWT and external identity token auth
- [OAuth / SSO Setup](https://docs.gizmosql.com/oauth_sso_setup.md): OAuth/SSO with external identity providers (Enterprise)
- [Bulk Ingestion](https://docs.gizmosql.com/bulk_ingestion.md): high-throughput data loading
- [DuckLake Integration](https://docs.gizmosql.com/ducklake.md): lakehouse catalogs with DuckLake
- [Geometry & Spatial](https://docs.gizmosql.com/geometry.md): spatial data support

## Features

- [Prometheus Metrics](https://docs.gizmosql.com/monitoring.md): independently licensed Enterprise HTTP endpoint and SQL metrics, configuration, metric reference, Prometheus/Grafana examples
- [Eager DDL/DML Execution](https://docs.gizmosql.com/eager_execution.md): no-fetch write semantics, prepared statement reuse, and client compatibility

- [SET Commands](https://docs.gizmosql.com/set_commands.md): runtime server/session settings
- [Session Instrumentation](https://docs.gizmosql.com/session_instrumentation.md): session and query observability (Enterprise)
- [Catalog Logging](https://docs.gizmosql.com/catalog_logging.md): fork server logs to an attached catalog (Enterprise)
- [Statement Queuing](https://docs.gizmosql.com/statement_queuing.md): admission control for concurrent statements (Enterprise)
- [Admin Command Gating](https://docs.gizmosql.com/admin_command_gating.md): dangerous instance-level commands (ATTACH/DETACH, SET GLOBAL, INSTALL/LOAD, local filesystem reads/COPY, duckdb_secrets) require the admin role — interim guardrail ahead of GizmoSQL 2.0 RBAC
- [Graceful Shutdown](https://docs.gizmosql.com/graceful_shutdown.md): drain in-flight queries on SIGINT/SIGTERM before stopping (for Kubernetes/containers)
- [OpenTelemetry Integration](https://docs.gizmosql.com/opentelemetry.md): traces and metrics export

## Integrations

- [Extensions & Tools](https://docs.gizmosql.com/integrations.md): ecosystem integrations
- [MCP Server](https://github.com/gizmodata/gizmosql-mcp): Model Context Protocol server and Claude Desktop extension so AI agents can explore schemas and run SQL against GizmoSQL
- [ADBC Scanner extension for DuckDB](https://docs.gizmosql.com/adbc_scanner_duckdb.md): query remote ADBC sources from DuckDB
- [ADBC extension for DuckDB (Columnar)](https://docs.gizmosql.com/adbc_duckdb_extension.md): profile-based ADBC reads and writes from DuckDB

## Optional

- [One Trillion Row Challenge](https://docs.gizmosql.com/one_trillion_row_challenge.md): benchmark write-up