DBConvert Streams

Database IDE + migration + real-time CDC — in one workflow.

Query data,
move it,
keep it in sync,
and let your AI assistant see it all

without switching between tools.

If this looks useful, consider giving it a ⭐

Why this exists

Most setups look like this:

• a DB client for queries
• scripts or tools for migration
• a separate CDC pipeline

It works, but it's fragmented.

DBConvert Streams combines these into one workspace.

What it feels like

Think of it as:

DBeaver / DataGrip

• migration tool
  
• CDC

but without switching tools every time

Example

Run queries across databases and files:

SELECT *
FROM read_parquet('orders.parquet') o
JOIN postgres.public.customers c
  ON o.customer_id = c.id
LIMIT 10;

Then use the same query as a data source — and stream it anywhere.

What you can do

• explore databases, files, and S3
• run SQL across multiple sources
• move data between systems
• keep it in sync with CDC
• connect Claude, Cursor, or Copilot — the AI reads your live schemas, data, and streams (read-only, via MCP)

All in the same workflow.

Note: This is the public home of DBConvert Streams — example configurations, documentation, issue tracking, and release notes. The product itself is proprietary.

Quick Start

Runs anywhere — your laptop, a VPS, or your own infra. No cloud dependency, no vendor lock-in.

Desktop App

Download for Windows, macOS, or Linux — no account required.

AI client extension

Grab dbconvert-streams-<version>.mcpb from Releases and drop it into Claude → Settings → Extensions. It asks for connection strings and folders, and nothing else has to be installed — no DBConvert Streams, no toolchain, no server to run.

postgres://user:password@host:5432/dbname
mysql://user:password@host:3306/dbname
s3://bucket/folder?region=us-east-1

Several sources go in the one field, separated by spaces — shop=postgres://… orders=mysql://… — and a single question can join across all of them.

Add folders of Parquet, CSV or JSON files with the folder picker. Every source becomes read-only tools in your chat, and one question can span several of them at once.

One file covers Windows and Linux — the bundle carries a binary for each and picks the right one. macOS is planned for a later release.

Not a Claude user? The same server runs as a container, and every MCP client that can launch one can use it:

docker run -i --rm slotix/stream-mcp "shop=postgres://user:password@host:5432/shop"

Cursor and VS Code can set that up in one click — Add to Cursor · Add to VS Code — VS Code asks for the connection string as you install it; in Cursor you replace the sample one. Folders, S3 keys and the rest: standalone server.

The bundled server is proprietary software, distributed under the DBConvert Streams licence. The MIT licence in this repository covers the examples, docs and assets here, not that binary.

Self-Hosted (Docker)

Deploy on any machine with Docker — a local server, a VPS (DigitalOcean, Hetzner, AWS EC2, etc.), or your own infrastructure:

curl -fsSL https://dbconvert.nyc3.digitaloceanspaces.com/downloads/streams/latest/docker-install.sh | sh

What is DBConvert Streams?

DBConvert Streams is a database IDE with built-in migration and real-time CDC.

Browse databases, local files, and S3 storage. Edit data directly. Run federated SQL queries that join tables across different database engines — no intermediate exports needed.

Key Features

In practice, it comes down to this:

Database IDE & Workspace (Free)

• Data Explorer — Browse databases, files, and S3 in one place
• ER Diagrams — Visualize database relationships
• Schema Comparison — Compare schemas and data across databases
• Schema Navigation — Persistent state and search across connections

Federated SQL

• Execute SQL queries across multiple databases and file sources simultaneously
• Join live PostgreSQL and MySQL tables using connection aliases
• Query CSV, JSON, Parquet files and S3 storage alongside databases

Built-in AI Chat — new in 2.5.0

Ask about the database work already open in the desktop app: a table, view, file, SQL console, connection, or migration/CDC stream. AI Chat starts with that live workspace context, so it can inspect schemas and data, explain or repair a failed query, and diagnose stream status without asking you to paste DDL into a separate chat.

Use your own installed agent CLI — Claude Code, Codex, GitHub Copilot CLI, or OpenCode. AI Chat automatically supplies the relevant scoped subset of DBConvert's read-only tools and shows tool activity while it works; it cannot change connections, configuration, streams, or data.

Docs: AI Chat →

AI Assistants via MCP — new in 2.4

• Built-in MCP server: Claude, Cursor, VS Code Copilot, Windsurf, Gemini CLI, and Codex read live schemas, data, and stream state — no more pasting DDL into chat
• 27 read-only tools: inspect workspace connections, schemas, tables and views; run read-only SQL and federated queries; compare schemas and samples; diagnose streams; browse files and S3
• One-click setup from the ✨ AI Assistants panel; Docker deployments expose /mcp over HTTP(S)
• Read-only by design: only SELECT passes the server-side filter — the AI can look and advise, never write

Docs: AI Assistants via MCP →

Tools

DBConvert Streams exposes these 27 read-only MCP tools. The names below match the live MCP server.

Connections and schema

• dbconvert_list_connections — list workspace connections
• dbconvert_get_connection — inspect one connection
• dbconvert_list_databases — list databases
• dbconvert_list_schemas — list schemas
• dbconvert_list_tables — list tables
• dbconvert_list_views — list views

Table and view inspection

• dbconvert_describe_table — inspect table columns and keys
• dbconvert_preview_table — preview table rows
• dbconvert_describe_view — inspect a view
• dbconvert_preview_view — preview view rows

Read-only SQL

• dbconvert_run_select — run a SELECT query
• dbconvert_explain_select — explain a SELECT query

Schema and data comparison

• dbconvert_compare_schemas — compare schemas
• dbconvert_compare_data_sample — compare data samples

Stream diagnostics

• dbconvert_list_streams — list streams
• dbconvert_get_stream — inspect a stream
• dbconvert_get_stream_status — get stream status
• dbconvert_get_stream_stats — get stream throughput and statistics
• dbconvert_get_stream_recent_errors — inspect recent stream errors
• dbconvert_get_stream_recent_logs — inspect recent stream logs

Files and S3

• dbconvert_list_files — list workspace files
• dbconvert_get_file_schema — inspect a file schema
• dbconvert_preview_file — preview file rows
• dbconvert_list_s3_buckets — list S3 buckets
• dbconvert_list_s3_objects — list S3 objects

Federated SQL

• dbconvert_run_federated_select — run a read-only query across sources
• dbconvert_explain_federated_select — explain a federated SELECT query

Only read-only operations are exposed: the server-side filter permits SELECT, so an AI client cannot alter connections, configuration, streams, or data. For client setup, see the MCP setup guide.

Data Migration (Load Mode)

Rapidly move large datasets between databases with automatic schema conversion and validation.

Performance: 23 million rows (4.38 GB) migrated from MySQL to Parquet in 35.7 seconds at 136 MB/s.

Real-time CDC (Change Data Capture)

Stream INSERT, UPDATE, and DELETE operations from source to target in real-time with minimal latency. Supports CDC to databases, files, and S3 storage.

When this is probably not for you

• you need 100+ connectors (SaaS, APIs, etc.)
• you already run Kafka pipelines at scale
• you need complex ETL / transformations

Screenshots

Data Explorer

Browse schemas, view and edit data across multiple database connections with a unified tree navigation:

Federated SQL

Join tables across MySQL, PostgreSQL, and file sources (CSV, Parquet) in a single query:

ER Diagrams

Visualize database relationships with interactive entity-relationship diagrams:

Stream Configuration

Configure data migration and CDC streams with table selection, custom queries, and transfer settings: