VisuaLeaf Quick Start Guide
This guide takes you from a fresh install to your first query result. It should take about fifteen minutes. You will install VisuaLeaf, create a connection to a database, explore its structure, run a query, and read the results.
VisuaLeaf works with MongoDB and with SQL databases — PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, and SQLite — from the same window. Where the steps differ between the two, this guide shows both.
Download and install VisuaLeaf for your platform. Builds are available for macOS (Apple Silicon and Intel), Windows, and Linux.
You also need a database to connect to. If you do not have one, MongoDB Atlas offers a free tier, or you can point VisuaLeaf at a local mongod, Postgres, or SQLite file.
When you launch VisuaLeaf, the Welcome screen opens. It shows your license status, quick actions for getting connected, and your three most recent connections.
Click Login or Register in the header. Registering starts your 14-day Pro trial — it runs from the moment you sign up — and unlocks the Quick Start actions, which stay disabled until a license is active.
Once you are signed in, the license card shows your email, plan, and expiration date. On a trial, a badge counts down the days remaining — blue above 7 days, orange at 4–7, red at 3 or fewer.
VisuaLeaf is organized around four areas. Everything in this guide happens in one of them.
A tree of your connections, databases, collections, and tables. This is where you navigate. See Sidebar.
Each thing you open — a collection, a query, an aggregation pipeline, a chart — becomes a tab. Tabs can be split into panels. See Layout System.
Press Ctrl + K (Cmd + K on macOS) to jump to any collection or command by typing. See Command Palette.
Along the bottom: connection state, row counts, query timings, and running background jobs. See Status Bar.
A connection stores everything VisuaLeaf needs to reach one server: the host and port, the database name, credentials, and any TLS or SSH settings. Connections are saved locally and can be grouped into projects and environments — for example a Billing project with dev, staging, and prod environments.
Open the Connection Manager from the Welcome screen or the toolbar, then click New Connection.
Billing — staging.mongodb:// and mongodb+srv:// Atlas strings work.PostgreSQL, MySQL, MariaDB, SQL Server, and Oracle follow the same shape:
5432 for PostgreSQL, 3306 for MySQL and MariaDB, 1433 for SQL Server, and 1521 for Oracle.SQLite has no server. Set Database Type to SQLite and point VisuaLeaf at the database file on disk. There is no host, port, or authentication to configure.
Click Test Connection. It validates the configuration without saving it, and shows step-by-step progress through network connectivity, authentication, the SSL/TLS handshake, and database access — so when something fails you can see which stage it failed at. When the test passes, click Save & Connect.
If your database is not reachable directly, open the SSH Tunnel tab and enter your jump host, port, and either a password or a private key. VisuaLeaf opens the tunnel and routes the connection through it — you do not need to run ssh -L in a separate terminal.
After connecting, your server appears in the sidebar as a tree. Expand it to work down through the hierarchy.
Two things worth learning early:
Click a collection in the sidebar. It opens in the Collection View, already showing the first page of documents.
{ status: "active" }.Enter or click Run.If you would rather not write the filter by hand, toggle the Query Builder and assemble conditions from dropdowns. The builder and the raw query stay in sync, so you can start visually and switch to text once you know what you want.
For SQL connections, open the SQL Editor from the activity bar or with Ctrl + Shift + S.
SELECT * FROM customers WHERE country = 'FR';. Autocomplete suggests table and column names as you type.F5 to run the whole buffer, or highlight a fragment and press F9 to run just the selection.;, you get one result tab per statement.Configure a provider under Settings → AI Assistant (OpenAI, Anthropic, Google Vertex AI, or a local Ollama endpoint), then describe your query in plain language — "find users who signed up this month with active subscriptions". Review the generated query and click Use Query to drop it into the query builder, pipeline builder, or shell. See AI Assistant.
Results are shown in whichever view suits the shape of your data. Switch between them at any time.
Ctrl + 1Documents as expandable nodes. Best for nested structures and arrays.
Ctrl + 2A spreadsheet grid, one row per document. Best for flat, uniform data.
Ctrl + 3Raw BSON with types spelled out. Best when types matter.
Ctrl + F.F5 to re-run the current query against the server.Click Explain in the toolbar to see the execution plan. Pick a verbosity — Query Planner, Execution Stats, or All Plans — and read whether the query used an index or scanned the collection. This is the fastest way to find out why something is slow. See Query Profiler.
Click any value in Tree or Table View to edit it inline. The input is type-aware — dates, ObjectIds, Decimal128, and arrays are validated as you type — and the change is saved as soon as you commit it. Every mutation goes through the command history, so Ctrl + Z undoes it and Ctrl + Shift + Z redoes it.
To get data out, right-click a collection or database in the sidebar and choose Export Data:
Long-running exports run in the background. Track them in the Task Manager.
On macOS, use Cmd wherever Ctrl is listed.
| Shortcut | Action |
|---|---|
Ctrl + K | Open the command palette |
Ctrl + F | Search within the current results |
Ctrl + 1 / 2 / 3 | Switch to Tree / Table / BSON view |
Ctrl + Shift + S | Open the SQL Editor |
F5 | Refresh results, or run the whole SQL buffer |
F9 | Run the selected SQL fragment |
Ctrl + F5 | Run SQL up to the cursor |
You now have a working connection and can read your data. These are the features most people reach for next:
Build multi-stage pipelines by dragging stages, with a live preview of the data at each step.
Turn a query or pipeline into a chart, then collect charts into a dashboard.
Review index usage, spot missing indexes, and create new ones without dropping to the shell.
See what your documents actually contain — field types, frequencies, and relationships — as a diagram.
A full mongosh-compatible shell against the same connection, for anything the GUI does not cover.
Copy or continuously sync data between clusters, with field-level control over what moves.
Look at which stage of the test failed. Stalling on network connectivity usually means a firewall or an Atlas IP allowlist; failing at authentication means the mechanism on the Authentication tab does not match what the server expects; failing at the TLS handshake points at the SSL/TLS tab.
Add your current IP to the Atlas Network Access allowlist, and make sure you pasted the mongodb+srv:// string with your real password substituted for the <password> placeholder.
Most features need an active license. Click Login or Register on the Welcome screen to activate your trial.
Run Explain on it. If the plan shows a collection scan, the fix is almost always an index — see Index Management.
Still stuck? Ask in Discord or send us a note from the contact page.
Download and start managing your MongoDB databases with ease.
Download Free Trial