🧭 Features

The full capability map of DataZen — from the AI assistant and MCP server to the SQL editor, data tools, and plugins.

🤖

AI Assistant

NL2SQL, diagnosis, EXPLAIN, Chat, smart filters

📊

Charts

Smart recommendations, five chart types, PNG/SVG export

⚙️

Workflow

Cross-database orchestration, AI-powered, reusable flows

🛰️

MCP Server / Client

Plug your databases into Claude, Cursor and other AI agents

🔌

Databases

PG / MySQL / SQLite / Redis / plugins

MCP Server & Client

Your databases, as tools for AI agents

DataZen includes a built-in MCP (Model Context Protocol) server. Point Claude Desktop, Cursor or any MCP-compatible agent at it, and they can inspect schemas, run queries, read EXPLAIN plans — even trigger whole workflows.

  • MCP server: expose database operations, schema inspection, EXPLAIN and workflows as MCP tools
  • MCP client: connect external MCP servers and call their tools from inside DataZen
  • Headless mode: run with --mcp-stdio on servers, scripts or CI — no UI required
  • Credentials stay local: agents talk to DataZen; passwords remain in your OS keychain
  • Persistent connections: agents address saved connections by ID instead of raw connection strings
Claude / Cursor / AI Agent
        │  natural language
        ▼
     MCP protocol
        │  tools · resources · prompts
        ▼
     DataZen (local)
        │  queries · schema · EXPLAIN · workflows
        ▼
PostgreSQL · MySQL · SQLite · Redis …

Setup guide: MCP server & client in the README

SQL Editor

Built for everyday queries

  • Syntax highlighting + table/column autocomplete
  • Multi-statement execution, query history, and favorites
  • SQL dialect auto-switches by connection type
  • Configurable result limits — large result sets won't freeze the UI
SQL editor
Data Browse & Edit

100K rows without lag

  • Virtual scrolling for large tables — smooth at 100K+ rows
  • Inline editing, safe batch delete, transaction support
  • Sort, filter, and pagination work together
  • Export CSV / JSON / SQL; import the same formats
Data browse and edit
Backup & Sync

Your data, under control

  • One-click SQL backup: Schema-only / Data-only / Gzip
  • Full app data export (including config), atomic writes + ZIP bomb protection
  • Data Sync — row-level diff within the same dialect family when structure & PK match
  • Data Transfer — heterogeneous one-way migration with a six-step wizard (structure and/or data)
  • Schema Diff — five-step wizard to compare structure and deploy controlled DDL (additive by default)
  • Backup packages automatically exclude key files
Data Sync, Transfer and Schema Diff
Security & Privacy

Secure by default

  • Connection passwords encrypted at rest; master key in system keychain
  • SSH credentials encrypted; host key verification on first connect
  • Hardened local IPC; file access restricted to trusted paths
  • SQL and natural-language requests not logged at info level by default
Security and privacy

Data capabilities at a glance

Capability Description
Virtual scrolling Smooth at 100K+ rows
Export / Import CSV, JSON, SQL interchange
Backup SQL dump (Schema / Data / Gzip)
Data Sync Same-family row-level diff when structure & PK match
Data Transfer Heterogeneous one-way migration (structure and/or data)
Schema Diff Structure compare + controlled DDL deploy
ER diagram Auto-generates table relationship graph
Redis view Key browser: String / Hash / List / Set / ZSet / Stream