Tamery：支持 MCP 的 AI 数据库客户端，可让 Claude Code、Cursor 查询数据库

The AI-powered database client and MCP server for desktop and web

We built Tamery because we wanted a database client that is fast, simple and nice to look at.

And one that actually helps instead of getting in the way.

It works with PostgreSQL, MySQL, MariaDB, SQL Server, ClickHouse and SQLite (Turso and Cloudflare D1 too).

Get started

brew install --cask tamery   # macOS
yay -S tamery-bin            # Arch Linux

Sign in. Or skip it on desktop with Continue without an account.
Paste a connection string. Or an .env file, a JDBC URL, a psql command. Tamery figures it out.
Open a table or run a query.

Tip

Start with the desktop app. It connects to localhost and private networks out of the box.

Table browser. Sort, filter, edit right in the cell.
Query editor with autocomplete that knows your schema.
AI assistant that writes, explains and fixes SQL. It never changes data without asking.
Schema visualizer.
Connections synced across devices, with or without passwords.
SSH tunnels and SSL.
MCP server. Claude Code, Cursor and other agents can query your databases. Every write waits for your click.

Anything that changes or deletes data asks first.
Works offline. Syncs when you are back.
Connection strings are always encrypted. With Metadata only they never leave your machine.
Everyday features are free. Pro adds the full AI assistant.
Open source. You can read every line that touches your credentials.

You need pnpm, Bun, and Postgres + Redis (or Docker).

pnpm install
pnpm run docker:start
pnpm run drizzle:migrate
pnpm run dev

The app opens at https://app.local.tamery.app. Tests: pnpm run test, and pnpm run test:e2e after pnpm run test:start.

We don't accept pull requests. But we read every issue.

Found a bug or something off by one pixel? Open an issue. Our perfectionists will thank you.

_{Built with ❤️}