gobang and SQLly: a terminal browser and a desktop workbench

gobang is a small MIT-licensed Rust TUI for MySQL, PostgreSQL and SQLite that has not shipped since 2021; SQLly is a desktop SQL workbench. Where each one fits.

By

gobang is a terminal database browser written in Rust. You run one binary, it draws a full-screen UI in your terminal, and you move through MySQL, PostgreSQL and SQLite databases with vim keys. It has about 3,300 stars on GitHub, it is MIT-licensed, and people still recommend it in “TUI tools for developers” lists. SQLly is also written in Rust, so the two get compared now and then, even though one lives in a terminal and the other is a desktop app.

We build SQLly, so take the bias into account. We read gobang’s README, its source on the main branch, its release assets and its commit history. We did not benchmark it. For some jobs it is the better tool, and we say which ones below.

First, the dates

gobang is dormant, and it’s fair to say so up front:

  • The newest release is v0.1.0-alpha.5, published on 15 September 2021. It is also the newest version on crates.io.
  • The last change to the code was a fix in December 2021.
  • The last commit of any kind was a README edit in January 2023.
  • The README still opens with “gobang is currently in alpha”, and an open issue from April 2024 asks for a new release with updated dependencies. There are 57 open issues.

None of that makes it useless. A small read-only browser for three databases doesn’t need much upkeep, and the MIT license means anyone can pick it up. But if you install it today, you are installing a tool as it stood in 2021.

Terminal against desktop

This is the real difference, and it decides most of the choice.

gobang runs wherever a shell runs. SSH into a server, start it, and you are looking at the database on that machine with nothing to tunnel or forward. It fits in a tmux pane next to your editor. Navigation is h/j/k/l, g and G, / to filter and y to copy a cell, and you never touch a mouse. Install is a one-liner through Homebrew (a tap), Scoop, nixpkgs, the AUR, NetBSD pkgsrc or cargo install. The release archives are 3.1 to 3.6 MB.

SQLly needs a graphical session. It is one Rust codebase drawn on the GPU through Metal, Direct3D or Vulkan, with downloads of 65.6 MB (macOS), 60.9 MB (Windows) and 76.2 MB (Linux). It reaches remote databases through SSH tunnels, proxies, or its own bridge and relay, but the app itself runs on your machine. It has a command-line tool, sqlly-cli, which runs an MCP server for AI agents. It has no terminal UI, and we’re not going to pretend otherwise.

Databases

EngineSQLlygobang
MySQL, PostgreSQL, SQLiteYesYes
SQL Server, Azure SQLYes (Azure Entra sign-in in progress)No
MariaDB, DuckDB, ClickHouse, Oracle, libSQL/TursoYesNo
Redis and its compatiblesYesNo
Wire-compatible services (CockroachDB, Neon, Supabase, Aurora…)Documented and testedNot documented

gobang talks to its three engines through sqlx 0.5. SQLly has 30 engines; the engine list spells out which tools work on each.

Connecting

gobang has no connection dialog. You edit ~/.config/gobang/config.toml (or %APPDATA%\gobang\config.toml on Windows) and add a [[conn]] block with type, host, port, user, database and an optional name. If you want it to log in without asking, the password goes in that file in plain text. There are no SSH tunnel or TLS settings in the config.

SQLly has a connection dialog with SSH hops, SOCKS and HTTP proxies, Kubernetes and TLS modes. Connections are plain files you can sync, but passwords stay in the OS keychain. It can also import connections from DataGrip, DBeaver, pgAdmin, TablePlus, Navicat and others.

What you can do once connected

In the released gobang you pick a database and a table from a tree, then flip between tabs for records, columns, constraints, foreign keys and indexes. You can filter records, select a block of cells with H/J/K/L and copy it. The grid is read-only.

The main branch goes a little further: an SQL editor tab (merged in October 2021), a properties tab, custom key bindings and named connections. None of it has been in a release, so you’d build from source with Cargo to get it, and the README still lists the SQL editor under TODOs. Its completion offers a fixed list of 12 keywords such as SELECT, WHERE and LIMIT.

SQLly starts where that stops. IntelliSense tracks aliases and scope, writes a JOIN ON clause from real foreign keys, pulls a column’s actual values into a WHERE clause, and flags a GROUP BY mistake as you type. The grid streams rows to disk past 50,000 rows, so 50 million rows use about as much memory as 50, and on top of it sit pivots, charts, conditional formatting, computed columns, result diffs and Excel export. A row editor built from the table’s metadata handles edits.

Safety and admin

gobang has no guardrails beyond the fact that its released grid can’t write. On main, the SQL tab will run whatever you type.

SQLly has guardrails for people typing SQL: SELECT-only connections that block file-reading and shell-escape tricks, a hard stop on UPDATE or DELETE without WHERE, automatic rollback, tenant checks, and production painted red. It also has the DBA tools a terminal browser doesn’t try to have: Activity Monitor, query plans (in progress across nine engines), schema and data compare, SQL Agent, an index analyzer and backup and restore as reviewable scripts.

What they cost

gobang is free and MIT-licensed, with no paid tier. SQLly has a free tier for hobbyists (two saved servers and a reminder to pay about every five hours), a one-time $100 SQLly 1.x license and a $200 Lifetime license; see pricing. SQLly is not open source. Lifetime licenses carry a source-escrow promise if SQLly is ever discontinued.

Which one to pick

Pick gobang if:

  • You work over SSH or in tmux and want to look at a table without leaving the terminal.
  • You want vim keys and no mouse.
  • You want a 3.5 MB binary from your package manager, not a 60-plus MB app.
  • You only need to browse MySQL, PostgreSQL or SQLite, and you’re fine with a tool that hasn’t changed since 2021.
  • You need free and open source at any company size.

Pick SQLly if:

  • You write SQL all day and want completions that know your schema and your data.
  • You use SQL Server, Azure SQL, DuckDB, ClickHouse, Oracle, libSQL or Redis.
  • You want to edit rows, pivot or chart results, or export to Excel.
  • You want guardrails on your own queries, plus DBA tools.
  • You want a tool that is still shipping releases.

Plenty of people will reasonably keep both: a terminal browser for the box you’re SSH’d into, and a desktop app for everything else. The side-by-side table is at SQLly vs gobang, and SQLly’s download is free.