The modern database engineering workbench

Your database.
Under control.

Name: DBNox
Price: 49 USD
Availability: InStock

Query data, explore schemas, compare environments, manage migrations, investigate performance and operate databases safely from one desktop engineering workspace.

Download DBNox Explore Features

Windows · macOS · Linux

Early bird $49 for the first 100 customers. Regular price $99. One year of updates.

DBNox · Commerce Platform

PRODUCTION

orders.sqlcustomers.sql

SELECT
  o.id,
  o.status,
  o.total,
  c.email
FROM orders o
JOIN customers c
  ON c.id = o.customer_id
WHERE o.store_id = 12
  AND o.created_at >= now() - interval '7 days'
ORDER BY o.created_at DESC
LIMIT 50;

Results · 50 rows

id	status	total	email
10482	paid	186.40	ada@shop.io
10481	paid	42.00	lin@shop.io
10480	pending	91.25	kai@shop.io

Supported databases

PostgreSQLMySQLMariaDBSQLiteSQL ServerMongoDBRedisClickHouseDuckDBand morePostgreSQLMySQLMariaDBSQLiteSQL ServerMongoDBRedisClickHouseDuckDBand more

Built with Rust + Tauri · Windows · macOS · Linux

Capability depth varies by engine.

Database work shouldn't require five different tools.

Writing SQL is only part of the job. You inspect schemas, edit data, analyze execution plans, compare environments, build migrations, investigate locks and move data between systems.

DBNox brings those workflows together in one database engineering workspace, so the context stays with the problem.

The usual stack

SQL ClientSchema ToolDiff ToolTerminalMonitoring

DBNox

Database

Query
Explore
Compare
Migrate
Diagnose
Monitor
Transfer
AI

Product overview

Everything you need to understand your database.

An SQL editor that understands your database.

Write SQL with schema-aware autocomplete, syntax highlighting and database-aware assistance. Run a statement, selection or complete script and keep results attached to the query that produced them.

Monaco editor with dialect awareness
Schema-aware autocomplete
Selected, current or full-script execution
Query tabs, history and saved queries
Split editor and results

orders.sql · Commerce Platform

STAGING

SELECT id, status, total

FROM orders

WHERE customer_id = cust|

Autocomplete

customer_id

created_at

customers

SQL IDE

An SQL editor that understands your database.

Write SQL with schema-aware autocomplete, syntax highlighting and database-aware assistance. Run a statement, selection or complete script and keep results attached to the query that produced them.

Monaco editor with dialect awareness
Schema-aware autocomplete from the connected database
Run selection, current statement or the full script
Multi-statement execution with cancellation
Query tabs, history, saved queries and SQL files

orders.sql · Commerce Platform

STAGING

SELECT id, status, total

FROM orders

WHERE customer_id = cust|

Autocomplete

customer_id

created_at

customers

Schema explorer

See the database, not just its tables.

Navigate databases, schemas, tables, columns, relationships, indexes, views and routines from one structured explorer.

Structured connection and object explorer
Inspector for columns, indexes, keys, constraints and DDL
ER diagrams for relationship context
Visual schema management without drowning in modals

customers · public

DEVELOPMENT

DataColumnsIndexesKeysConstraintsTriggersDDLStatistics

Column	Type	Keys
id	uuid	PK
email	text
status	varchar(20)
created_at	timestamptz

Data grid

Work directly with your data.

Browse large result sets without loading everything into memory. Filter, sort, inspect and safely edit records from a virtualized data grid.

Virtualized grid with server-side pagination
Sorting, filtering and column management
Pending-change review before commit
Row inspector, JSON/JSONB and FK navigation

orders · pending edits

PRODUCTION

3 changes pending

Review ChangesCommitDiscard

id	status	total	store_id
10482	shipped	186.40	12
10481	paid → refunded	42.00	12
10480	pending	91.25	8

Query profiler

Find out why your query is slow.

Turn complex execution plans into something you can investigate. DBNox visualizes query plans, highlights expensive operations and connects findings back to the tables and indexes involved.

Plan visualization with expensive-operator highlights
Findings derived from the database and execution plan first
AI can explain findings and prepare reviewable SQL second
No fabricated improvement percentages

QUERY PROFILER · orders lookup

PRODUCTION

Execution: 4.82s
Rows scanned: 2.8M
Returned: 38

Limit
Sort 1.2s
Hash Join
Sequential Scan · orders 2.8M rows

Findings

Sequential scan on orders
2.8M rows scanned for 38 results
Sort spilled to disk

Potential indexorders(customer_id, created_at)

View EvidenceOpen SQL

Production safety

Production should look like production.

Classify connections as local, development, staging or production. DBNox applies a central Safety Engine across the SQL editor, data grid, AI, MCP, compare, migrations and transfer workflows.

Environment-aware policies

Production can stay read-only by default while development allows faster iteration with explicit confirmation.

Dangerous-query detection

Unbounded UPDATE and DELETE, DROP, TRUNCATE and other high-impact statements are classified before they execute.

Protected destructive operations

Mutations can require review, typed confirmation, affected-row preview or a temporary safety token — not a silent run.

Auditable changes

Reasoned, reviewable changes instead of accidental writes across editor, grid and agent workflows.

Production Safety

PRODUCTION

● PRODUCTION

DELETE FROM orders;

Blocked by Production Safety

DELETE has no WHERE clause.

CancelReview Query

DBNox Intelligence

Preview

AI that investigates with evidence, not guesses.

DBNox AI writes and explains queries in the editor, reads execution plans, and can investigate with typed database tools. Writes wait for your approval. A local model can stay on the machine.

Generate in the editor
AI Assist and /ai turn a plain-language request into SQL, or into a Redis command on a Redis connection. Search engines are asked in that engine's query form.
Explain a query
Explain the statement under the cursor, or a notebook cell, without executing it.
Read an execution plan
Visual EXPLAIN can send the plan to the model and show the analysis beside the operators.
Investigate with tools
The investigator can call EXPLAIN, statistics, indexes, locks and sessions, then explain what those tools returned.
Suggest a container command
Describe what you want inside a container. DBNox asks the model for one shell command and shows the prompt before you run it.
Local models
Run Qwen through the bundled llama.cpp server, and start it when DBNox opens, or use Ollama. Cloud providers stay optional.

DBNox Agent

PRODUCTION

Why is this query taking 12 seconds?

Investigating

✓ Inspected query
✓ Loaded execution plan
✓ Checked indexes
✓ Checked table statistics
✓ Checked lock state

Finding

orders scans 8.7M rows while returning 214.

No existing index covers store_id + status + created_at

View EvidenceOpen SQLCreate Migration

AI ACCESS

PRODUCTION

Read schema✓
Read metadata✓
Execute SELECT✓
Run EXPLAIN✓
INSERT○
UPDATE○
DELETE○
ALTER○
DROP○

Production mode: Advisory only

You decide what AI can do.

Keep AI advisory, allow read-only investigation, or approve each write. AI activity records what was asked. Per-connection notes can carry schema conventions the assistant should follow.

Your model. Your choice.

API keys stay in the operating-system keychain. A local Llama or Ollama runtime needs no cloud key. Choosing a model does not grant it database permissions.

OpenAIAnthropicMiniMaxOpenRouterOllamaLocal LlamaOpenAI-compatible

Client, engineering, intelligence and operations — one workspace.

DBNox is not a pile of disconnected utilities. Compare, migrate, monitor, AI and MCP share connections, safety policies, jobs and audit infrastructure.

Database client

Daily query, explore and edit workflows in one workspace.

Queries become investigations.

Combine SQL, Markdown, parameters, results and charts in reusable notebooks for analysis, troubleshooting and documentation.

Edit and compare XML.

Open XML from files or grid cells. Source and tree views, with a structural compare that ignores indentation.

Inspect JSON like a first-class datatype.

Open JSON and JSONB from files, query results or database cells. Tree, table, form and source views with diff, patch and schema inference.

▸ Document and compare modes
▸ Lossless diff and JSON Patch
▸ Send to Compare from grid cells
▸ Snapshots for before/after review

Query results you can reuse.

Save, export, chart and compare result sets instead of treating every grid as disposable output.

▸ Save results as workspace artifacts
▸ Compare two saved results or queries
▸ Export to CSV and JSON

Coming Soon

Find anything across the workspace.

Search databases, schemas, tables, columns, views, routines, saved queries and snippets from one index — even across hundreds of schemas.

Database engineering

Compare, migrate, transfer and verify environment changes.

Know exactly what changed.

Compare databases, schemas or saved snapshots. See structural differences and generate reviewable migration SQL before deployment surprises.

▸ Side-by-side schema diffs
▸ Destructive changes called out clearly
▸ Select individual changes to synchronize

Compare more than schemas.

Find records that are missing, changed or different between environments, then inspect field-level differences before synchronizing.

▸ Primary, unique or custom key mapping
▸ Column mapping across connections
▸ Preview INSERT, UPDATE and DELETE sync SQL

Database changes you can review.

Create and apply versioned migrations with checksums, execution history and explicit rollback SQL when available.

▸ Schema Compare → Migration → Review → Apply
▸ Preflight and safety analysis
▸ Migration records with verification

Move data without building another script.

Import, export and copy data between supported databases with mapping, type conversion, validation and progress tracking.

Coming Soon

Portable database snapshots.

Capture schema metadata, DDL, relationships and selected statistics for offline inspection, drift baselines and support diagnostics.

Coming Soon

Catch environment drift early.

Persist production ↔ staging relationships, compare schema snapshots on demand and review drift history before it becomes an incident.

Coming Soon

Refresh dev without leaking production.

Copy production into development with masking, deterministic transforms and referential consistency across related tables.

Coming Soon

Manage roles, grants and ownership across schemas, tables and functions with SQL preview before execution.

Coming Soon

Explain effective permissions.

Answer why a user can SELECT a table by tracing role membership, grants, PUBLIC, ownership and row-level security.

Operations

Monitor live systems, run background work and connect agents.

See what's happening right now.

Investigate sessions, active queries, locks, waits, table statistics, indexes and database health without leaving the workspace.

▸ Sessions, locks and long-running queries
▸ Index inventory and usage stats
▸ Engine-specific diagnostics

Long-running work belongs in the background.

Keep working while DBNox handles imports, exports, comparisons, backups and transfers. Track progress, cancel operations and inspect failures.

Preview

Built for humans. Ready for agents.

Expose selected connections to MCP clients with per-connection policies — schema only, read only, read + write or custom — enforced at the DBNox tool layer.

▸ Hard enforcement of read/write policy
▸ Audit log of agent tool calls
▸ Same safety architecture as internal AI

Local terminal and SSH shells.

Open a local shell or an SSH session next to the database you are working on.

Files: SFTP, FTP, FTPS and S3.

Browse local files, SFTP from an SSH connection, and FTP, FTPS or S3 profiles, then transfer between panes.

Docker and Podman containers.

List and inspect Docker and Podman containers, and run a command inside one. DBNox does not claim a Kubernetes GUI.

REST and HTTP API client.

Keep REST and HTTP endpoint profiles in the same workbench. DBNox does not claim gRPC or GraphQL parity.

Coming Soon

Headless CLI for CI and automation.

Run schema diff, data compare, transfer and snapshot commands from scripts with JSON output and CI-friendly exit codes.

Desktop-native. Local-first.

Rust-powered core

Database, filesystem and system operations run through the native DBNox backend.

Credentials stay protected

Store supported credentials through operating-system secure storage rather than plaintext application configuration.

Direct database connections

DBNox connects from the desktop without requiring a mandatory DBNox database proxy.

Local AI available

Use supported local models when database context should remain in your environment.

One DBNox. Every desktop.

Built with Tauri, Rust and modern web technologies for a lightweight cross-platform desktop experience.

Windows

macOS

Linux

More than a SQL client.

A factual capability comparison. Unreleased DBNox workflows are marked instead of implied.

Capability	Typical SQL client	DBNox
SQL editing	Yes	Yes
Data editing	Yes	Yes
Schema exploration	Yes	Yes
JSON and XML editors	Varies	Yes
Terminal and file transfer	Varies	Yes
Docker and Podman	Varies	Yes
REST API client	Varies	Yes
Visual EXPLAIN	Varies	Yes
Schema comparison	Varies	Yes
Data comparison	Varies	Yes
Migrations	Varies	Yes
Import / export / transfer	Varies	Yes
Live monitoring	Varies	Yes
Query diagnostics	Varies	Yes
Production safeguards	Varies	Yes
Permissioned AI	Varies	Preview
MCP for agents	Varies	Preview
Notebooks	Varies	Yes
SSH / SFTP	Varies	Yes
Background jobs	Varies	Yes
Environment drift detection	Varies	Coming Soon
Data sanitization pipelines	Varies	Coming Soon

DBNox vs DBeaver leads with memory: DBNox is about 15 MB. DBNox vs Mongrel covers licensing.

DBNox vs DBeaver DBNox vs Mongrel

Questions, answered precisely.

What is DBNox?+

DBNox is a desktop workbench for querying data, exploring schemas, comparing environments, migrations, monitoring and JSON and XML editing. It also includes a terminal, file transfer (SFTP, FTP/FTPS and S3), Docker and Podman containers, and a REST API client. Production safety and permissioned AI are included; a Kubernetes GUI and gRPC or GraphQL clients are not.

Which databases does DBNox support?+

Validated engines currently include PostgreSQL, MySQL, MariaDB, SQLite, MongoDB, Redis, ClickHouse and DuckDB, with SQL Server in preview. Capability depth varies by engine and feature.

Does DBNox run on Windows, macOS and Linux?+

Yes. DBNox is built with Tauri and Rust for Windows, macOS and Linux. Windows is available now: 64-bit .exe and .msi installers on the download page, and the Microsoft Store listing at https://apps.microsoft.com/detail/xpfnwc91lzswht. macOS and Linux builds are not published yet.

Can DBNox compare schemas and data between environments?+

Yes. Schema Compare diffs databases, schemas or snapshots and generates reviewable migration SQL. Data Compare finds missing, changed and equal rows between tables or query results with configurable keys and column mapping.

How does DBNox handle production safety?+

Connections can be classified by environment. A central Safety Engine applies policies across the SQL editor, data grid, AI, MCP, compare and migration workflows — blocking or requiring confirmation for dangerous statements like unbounded DELETE or DROP on production.

Does DBNox support AI?+

Yes, as Preview. In the editor, AI Assist and /ai generate SQL or a Redis command and can explain the statement under the cursor. Visual EXPLAIN can analyze a plan. The investigator uses EXPLAIN, statistics, indexes and locks before it explains a finding. Container exec can ask for one shell command. Providers include OpenAI, Anthropic, MiniMax, OpenRouter, Ollama, a bundled local Llama runtime, and OpenAI-compatible endpoints. Writes wait for approval.

What is MCP in DBNox?+

Model Context Protocol lets compatible AI development tools access selected DBNox connections through a Safe MCP Gateway. Each connection can be schema-only, read-only, read + write or custom. Policies are enforced at the DBNox tool layer with audit logging.

Can DBNox move or mask data between databases?+

Import, export and database-to-database copy are available today. Reusable sanitization pipelines that mask production data for development refresh are on the roadmap.

Does DBNox send my database through a cloud service?+

No. DBNox connects to your databases directly from the desktop and stores supported credentials in operating-system secure storage. Optional cloud AI only receives the context you explicitly allow — using a cloud model is not the same as keeping database context local.

Is DBNox a subscription?+

No. You pay once per seat. Early bird is $49 USD for the first 100 customers. Regular price is $99 USD. The seat includes 12 months of major and minor updates and patches. After that year, the app keeps working and new updates stop.

How does DBNox compare to DBeaver?+

DBNox’s memory footprint is about 15 MB. DBeaver is a Java client with a default maximum heap of 1 GB and a much wider set of databases. DBNox also includes compare, migrations, JSON and XML editors, a terminal, files, Docker and Podman, and a REST client. Early bird is $49 USD per seat for the first 100 customers. Regular price is $99 USD, paid once. See the comparison page.

How does DBNox compare to Mongrel?+

Both are desktop workbenches. DBNox combines database engineering — SQL, schema and data compare, migrations, monitoring, JSON and XML editors — with a terminal, files (SFTP, FTP/FTPS and S3), Docker and Podman containers, and a REST API client. Early bird is $49 USD per seat for the first 100 customers. Regular price is $99 USD, paid once, with 12 months of updates. The app keeps working after that. Mongrel, from VisorCraft, is billed annually and markets a wider engine list plus gRPC/GraphQL and Kubernetes depth DBNox does not claim. See the dated comparison page.

Where can I download DBNox?+

Windows builds are on https://dbnox.com/download as a 64-bit .exe and .msi, and on the Microsoft Store at https://apps.microsoft.com/detail/xpfnwc91lzswht. macOS and Linux builds are not published yet.

What do Released, Preview and Coming Soon mean?+

Released features ship in current desktop builds. Preview features are usable but still evolving. Coming Soon features are on the public roadmap and not yet available — the site labels them explicitly rather than implying they exist today.