DOCS

From sign-up to your first migration.

CanvaDb is a visual canvas for mapping, querying and migrating databases — without raw SQL or ops scripts. Everything you need is on this page.

GET STARTED

Up and running in four steps.

Create an account, verify it and set up a workspace. Then connect your first database.

  1. 01

    Create an account

    Sign up with your email, or continue with Google, GitHub or Microsoft. The free plan needs no card.

  2. 02

    Verify your email

    Open the link we email you to confirm your address — you can’t sign in until you do. It expires after 24 hours; use resend if needed, and check spam.

  3. 03

    Complete onboarding

    A short four-step wizard asks about your organisation, the databases you use, your migration goals and your team. You can skip it and come back later.

  4. 04

    Create a workspace

    From Projects, create a workspace — one canvas per environment, like “Production” or “Staging”. Free accounts have one workspace.

CONNECT

Connect a database.

Add a database to the canvas and choose Connect. Paste a connection string — CanvaDb parses it, tests the connection and pulls your schema. A read-only database user is all it needs.

Credentials are encrypted at rest with AES-256-GCM and never stored in plain text. Query results are never stored.

Connection string formats

PostgreSQLpostgresql://USER:PASSWORD@HOST:5432/DATABASEpostgresql://readonly:••••@db.acme.internal:5432/orders

Add ?sslmode=require for cloud hosts such as RDS, Supabase or Neon.

MySQLmysql://USER:PASSWORD@HOST:3306/DATABASEmysql://readonly:••••@db.acme.internal:3306/shop

Add ?ssl=true for PlanetScale, AWS RDS or any host that requires TLS. MariaDB 10.3+ works too.

MongoDBmongodb+srv://USER:PASSWORD@cluster.mongodb.net/DATABASEmongodb+srv://readonly:••••@cluster0.abc12.mongodb.net/events

Use mongodb+srv:// for Atlas, or mongodb://localhost:27017/dbname locally.

SQL Servermssql://USER:PASSWORD@HOST:1433/DATABASEmssql://readonly:••••@sql.acme.internal:1433/master

Use SQL Server authentication (username and password); Windows auth isn’t supported.

DynamoDB

DynamoDB uses AWS credentials instead of a connection string: your region, access key ID and secret access key.

Where to find your connection string

Supabase
Project Settings → Database → Connection string (URI)
Neon
Dashboard → your project → Connection details
PlanetScale
Connect → pick a framework → copy DATABASE_URL
MongoDB Atlas
Database → Connect → Drivers → copy the SRV string
AWS RDS
RDS console → your database → Connectivity → endpoint and port
MAP AND QUERY

See every table. Ask it anything.

Connected databases appear as nodes on the canvas. Drag them around, draw edges to map how systems relate, then open a node to work with its tables.

  1. 01

    Open a database

    Double-click a node to open the builder: a schema explorer on the left, a table canvas with primary and foreign keys in the middle, and a query builder.

  2. 02

    Edit the schema

    Add tables, columns, keys and relations visually. Types stay native to each engine. Export the schema as DDL for Postgres, MySQL, SQL Server, MongoDB or DynamoDB.

  3. 03

    Run queries

    Write SQL or Mongo queries and run them against the live connection (⌘↵). Explain shows the plan without executing (Postgres and MySQL). Destructive queries ask for confirmation first.

  4. 04

    Enhance with AI

    Describe what you want in plain English, or paste a query to optimise. The AI sees your schema, never your rows, and its change is shown as a diff to accept or reject.

Free accounts get 20 AI uses in total; Builder and Enterprise are unlimited. Export results to CSV or Excel.

MIGRATE

Move data between engines.

Pipelines extract from the source in batches of 500 rows, convert them to a consistent format and load them into the target 100 rows at a time. Progress is counted as data is extracted, so it always moves.

  1. 01

    Create a pipeline

    Open Pipelines and choose New pipeline. Name it, then pick a source and a target: MongoDB, PostgreSQL or MySQL, in any direction.

  2. 02

    Add connection strings and start

    Paste the source and target connection strings and start the run. Strings are encrypted with AES-256-GCM; prefer a read-only replica as the source.

  3. 03

    Watch the live log

    Every stage streams in as it happens — connection, schema mapping, extraction, table preparation and batch loads — with INFO, SUCCESS, WARNING and ERROR levels. The page updates every 3 seconds.

  4. 04

    Share or re-run

    When the status reaches COMPLETED, share a read-only link to the run, or start it again. Re-runs rebuild the target tables, so there are no duplicates.

Pipelines are on every plan. Free accounts get one pipeline and up to 1 GB of transfer; Builder and Enterprise are unlimited.

Supported source → target
Source ↓ / Target →MongoDBPostgreSQLMySQL
MongoDBSupportedSupportedSupported
PostgreSQLSupportedSupportedSupported
MySQLSupportedSupportedSupported

SQL Server and DynamoDB can be connected and mapped, but aren’t pipeline sources or targets yet.

Common questions

What happens to existing data in the target?

For PostgreSQL and MySQL targets, each table is dropped and recreated before its first batch, so every run starts clean. MongoDB targets upsert documents by _id.

Are column types preserved?

Not yet. SQL target columns are created as TEXT, and dates, numbers and booleans are written as strings. Native type mapping is on the roadmap.

How does the 1 GB free limit work?

Free accounts can move up to 1 GB in total across their pipelines. Bytes are counted as they are extracted; when a run crosses the limit it stops and logs why. Builder and Enterprise have no transfer limit.

Can I share a run with someone outside my team?

Yes. From the run page, choose Share and turn the link on. Anyone with it sees status, progress and the log, read-only and without signing in. Credentials are never shown.

CONNECTORS

Five engines today. Six on the way.

CanvaDb connects straight to your database to pull its live schema. No data leaves your database unless you run a migration.

  • Live
    PostgreSQL12 – 16

    Schema pull from information_schema, live queries and migrations.

  • Live
    MySQL5.7, 8.x

    Schema pull from INFORMATION_SCHEMA, live queries and migrations. MariaDB 10.3+ supported.

  • Live
    MongoDB4.x – 7.x

    Collections and inferred schema from Atlas or self-hosted, shell-syntax queries and migrations.

  • Beta
    SQL Server2017 – 2022, Azure SQL

    Connection testing and mapping on the canvas. Full schema pull and pipelines are in development.

  • Beta
    DynamoDBAWS, all regions

    Connect and map on the canvas. Schema introspection and queries are in development.

Coming soonroadmap · subject to change
  • Redis / ValkeyQ4 2026Key inspection, TTL viewer and key-pattern maps.
  • Cassandra / ScyllaDBQ4 2026CQL schema introspection and keyspace mapping.
  • SnowflakeQ1 2027Warehouse schema explorer and query builder.
  • BigQueryQ1 2027Datasets and tables with partitioning and clustering metadata.
  • ClickHouseQ2 2027Columnar schema explorer with MergeTree engine metadata.
  • Elasticsearch / OpenSearchQ3 2027Index mappings and a query DSL builder.
TEAM AND SHARING

Bring your team in.

Comment on any table, see who else is on the canvas, and share read-only links with people who don’t need an account.

  1. 01

    Invite teammates

    On the Organisation page, create departments and invite people by email. Give each person Read, Write or Admin access per workspace.

  2. 02

    Share a read-only link

    From the canvas, choose Share and turn on the public link. It shows layout and schema only — never credentials — and you can revoke it any time.

  3. 03

    Comment and collaborate

    Leave comments on any database or table. On Builder, see teammates’ cursors and selections live.

  4. 04

    Keep an audit trail

    Builder and Enterprise workspaces log who changed what in the activity panel.

Member invites, live presence and audit logs need Builder or Enterprise. Read-only links and comments are on every plan.

Ready to map your databases?

The free plan includes one workspace, one connection, 20 AI uses and pipelines up to 1 GB. No card needed.