Basevolt vs SQLly: an admin panel vs a SQL editor
Basevolt is an Electron admin panel with kanban, dashboards, webhooks and MCP. Here's how it compares to SQLly on engines, SQL, safety and cost.
Basevolt calls itself “the admin panel you use after AI built the app.” You point it at a PostgreSQL, MySQL, SQLite or Cloudflare D1 database and it gives you grids, kanban boards, gallery cards, dashboards and webhooks on top of the tables you already have. It’s made by one independent developer, Dylan Yu, who says he runs nine of his own apps with it. Its blog also writes a lot of “alternatives” posts (TablePlus, DBeaver, pgAdmin, phpMyAdmin) and lists itself in them, which is how it ended up on my radar. So I put the two side by side.
I build SQLly, so I’m biased. Here’s what I actually checked: Basevolt’s site, pricing page, FAQ, MCP page and changelog, its update feeds and file sizes, and the 0.16.0 macOS app bundle, which I opened up and read (it’s not open source, so there’s no repo to read). I didn’t run it against a database and I didn’t benchmark it.
The short answer up front: these two barely compete. Basevolt is for looking after an app’s data without writing SQL. SQLly is for writing SQL. I couldn’t find a SQL editor in Basevolt at all, on the site or in the bundle. If that’s what you need, you can stop here. If you’re not sure, read on.
How they’re built
- Basevolt is an Electron app. The front end is React with TanStack, and a small Node server runs inside the app and talks to your database through the usual Node drivers: pg, mysql2 and better-sqlite3. It updates itself through electron-updater.
- SQLly has no webview. The engines, parser, IntelliSense and tools are a Rust core, and the UI is one Rust codebase drawn on the GPU through Metal, Direct3D or Vulkan.
That shows up in the download. Basevolt 0.16.0 is 124.7 MB for the Apple Silicon dmg, 96.5 MB for the Windows x64 installer, 90.1 MB for the .deb and 136.4 MB for the AppImage. SQLly’s macOS dmg is 65.6 MB. I usually lose this one, so it’s nice to win it for once.
Platforms are closer. Basevolt is signed and notarized on macOS, with separate Intel and Apple Silicon builds, Windows installers for x64 and arm64, and Linux for x86_64 only. SQLly is signed and notarized on macOS but Apple Silicon only, and its Windows and Linux builds are a supported beta on amd64 and arm64. If you’re on an Intel Mac, Basevolt runs and SQLly doesn’t.
Databases
| Engine | SQLly | Basevolt |
|---|---|---|
| PostgreSQL | Yes | Yes |
| MySQL | Yes | Yes |
| SQLite (local files) | Yes | Yes |
| Supabase and other Postgres-compatible services | Yes | Yes |
| Cloudflare D1 | No | Yes |
| Remote SQLite over HTTP (sqlited) | No | Yes |
| MariaDB | Yes | Not listed |
| SQL Server | Yes | No |
| DuckDB, ClickHouse, Oracle | Yes | No |
| libSQL/Turso | Yes | No |
| Redis | Yes | No |
SQLly covers 30 engines if you count the documented wire-compatible services. Basevolt covers four plus the services that speak their protocols. That’s fine for what it’s for: most apps people build with Claude or Cursor land on Postgres, MySQL or SQLite anyway. D1 is the one where Basevolt has something I don’t. SQLly can’t open a D1 database today.
What you actually do in each
Basevolt: views over your tables
Basevolt reads your schema without migrations or config files and builds what it calls a base. From there you get a grid, a kanban board, a gallery and dashboards with charts, metrics and tables. You can rename fields and format values for display only, so your real schema stays untouched. Record pages open their related records, even when the foreign keys were never declared. Action buttons call webhooks, and there’s a delivery log. Version 0.16.0, out November 9, added inline grid editing and CSV import.
That’s a back office. If you sell something and someone on your team needs to see unprocessed refunds every morning and click a button that hits your API, Basevolt does that and SQLly doesn’t. SQLly has no kanban, no gallery and no webhooks.
SQLly: the SQL and the work around it
SQLly is built around the editor. Completions pull a column’s real distinct values into a
WHERE, write a join’s ON clause from your foreign keys, and catch
a GROUP BY mistake as you type. Results stream to disk past 50,000 rows or 64 MiB, so
a 50-million-row result stays flat on memory. Then there’s pivot and charts on any result,
CSV, JSON and Excel import and export, schema and data compare, an activity monitor, and backup
and restore as scripts you read before they run. The plan viewer is still in progress, so
don’t pick SQLly for that alone yet.
Basevolt’s data in and out is CSV import into a table and CSV export of a view. It has DataMap, an auto-laid-out diagram of your tables, so ER diagrams are a tie.
Connections and safety
This is where I’d slow down if your database isn’t on your laptop. I found no SSH tunnel option on Basevolt’s site and no SSH library in the 0.16.0 bundle. There’s an SSL switch, but in the code I read it turns off certificate verification when it’s on. For a hosted Postgres that’s still encrypted, just not checked. SQLly has SSH hops, SOCKS and HTTP proxies, and TLS modes.
Passwords: Basevolt’s FAQ says credentials are encrypted and stored locally, and the bundle does use AES in its own store. I didn’t find keychain integration. SQLly keeps connection files as plain text you can commit, and keeps the passwords in the OS keychain.
Guardrails: SQLly has SELECT-only connections, a hard stop on UPDATE or
DELETE without a WHERE, auto-rollback and prod painted red. I
couldn’t confirm a read-only connection mode in Basevolt.
AI and MCP
Both have an MCP server, and they made opposite choices. Basevolt’s is local, comes with
every plan, and has about 37 tools. Most of them configure the panel: views, fields, dashboards,
webhooks. But it also has create, update and delete for records, and a run_sql tool
that only accepts SELECT-style statements, enforced by checking keywords. So an agent
can change your rows through it. That’s the point for some people.
SQLly’s sqlly-cli mcp lets an agent read schemas and run one SELECT
in a read-only transaction that’s always rolled back. Anything that would write comes back
to you as a proposal. I’m paranoid about agents and production data, and that’s the
design I’d want.
For chat, Basevolt has a built-in assistant that queries data and builds dashboards with your own Claude, OpenAI or Gemini key, and a one-click setup that needs no key. SQLly works with local models (Ollama, MLX), your own endpoint, or opt-in cloud providers.
One more thing I noticed in the bundle: Basevolt includes the Aptabase analytics SDK and sends an
app_started event with the version and platform. That’s light, but the privacy
policy doesn’t mention it. SQLly has no telemetry today.
Cost
- Basevolt Hobby is free with no account: two projects on one machine, with views, dashboards, webhooks and MCP included.
- Basevolt Pro is $99 a year: unlimited projects and data sources, sync across up to three machines, and if you stop paying you keep the version you had.
- SQLly is free for hobbyists with two saved servers. Past that it’s a one-time $100 for SQLly 1.x or $200 for Lifetime.
Over two years Basevolt Pro costs $198, about the same as SQLly Lifetime. Basevolt’s fallback license is a fair deal for a subscription, and I like it.
Which one to pick
Pick Basevolt if:
- You want an admin panel for an app, with kanban, gallery and dashboards, and you’d rather not write SQL.
- You want action buttons and webhooks on your records.
- Your app runs on Cloudflare D1 or remote SQLite.
- You want an AI agent to set the panel up for you over MCP.
- You’re on an Intel Mac.
Pick SQLly if:
- You write SQL every day and want an editor that knows your schema and your data.
- You connect through SSH, or you work with SQL Server, MariaDB, DuckDB, ClickHouse, Oracle or Redis.
- You want read-only connections and an MCP server that can’t change data.
- You need schema compare, activity monitoring or backup and restore.
- You’d rather pay once than every year.
And for what it’s worth, they don’t fight each other. Basevolt for the people who run the app, SQLly for the person who writes the queries, is a reasonable setup. Basevolt is at 0.16 from one developer and SQLly is closed source from a small shop, so neither one is a sure bet forever.
The full feature table is at SQLly vs Basevolt. If you want to try SQLly, it’s free to download, and I’d like to hear what you think of it.