Introduction
Lizard is an AI-native, zero-config data console. Point it at one or many databases — mixing engines freely — and get an auto-generated CRUD UI, natural-language querying, and prompt-driven charts across the whole fleet, with zero manual admin-panel configuration.
What it does
- Auto-generated CRUD UI — browse, filter, and edit every table across every schema and database, derived entirely from schema introspection.
- Natural-language querying — an on-demand agent explores only the tables it needs, streams its progress live, and can be stopped mid-run. The generated SQL is always shown and editable.
- One-click & prompt-driven charts — visualize any result, or describe a chart in words. A single chart can join data from different microservice databases.
- Dashboards — a saved result is a page of blocks, each one query plus how to read it: table, chart, kanban, or gallery, with variables and an optional date range in the URL.
- Bounded by design — no unscoped catalog, no query without a limit, no cache or connection pool that only grows. Everything has a ceiling, because the target is a database per microservice.
Why a fleet, not a database
Most teams run microservices, so their data is deliberately split across many databases — one per service. Lizard treats that fleet as a single queryable surface: reads and analytics can span databases and engines via an embedded, read-only federation engine, while writes always target exactly one owning database.
Where to go next
- Features — the full tour, including filters, timezones, dashboards, and the limits that keep it bounded.
- Quickstart — run Lizard locally or in Docker in a few minutes.
- Architecture — how the console, the guard, and federation fit together.
- Security model — the two-role model, the SQL Guard, and credential encryption.
- Roadmap — what's shipped, what's in progress, and what's deliberately out of scope.
- FAQ — common questions before you self-host.
- Source on GitHub — file issues, read the code, contribute.