Help

AddisDB help center

How to connect, query, ship changes safely, and get the most out of the app.

Getting started

  • Installing AddisDB — Download the right build for macOS, Windows or Linux, get past the first-launch prompt, and connect your first database.
  • AddisDB for VS Code — Install the extension from the Marketplace, and browse and query the databases you set up in the desktop app without leaving your editor.
  • The addisdb command line — Install the addisdb CLI with one curl command, get a full terminal UI for your databases, and call the same work from a script or CI.
  • Writing and running SQL — Schema-aware autocomplete, running just the statement you are on, saving your work as .sql tabs, and getting unstuck when a query fails.
  • Keyboard shortcuts — The handful of keys worth knowing. They are the same on every platform — ⌘ on macOS, Ctrl on Windows and Linux.
  • Do I need an account? — Not for the databases on your own machine — those are free forever. Here is what a plan does cover, and what stays on your machine either way.

Connections

Database guides

  • PostgreSQL — The engine AddisDB supports most deeply — schema diagrams, in-grid editing, live monitoring, throwaway test clones, and AI, all wired up.
  • MySQL — Connect to MySQL and get the full workspace — schema diagram, in-grid editing, live monitoring from the performance schema, mock data and AI.
  • MariaDB — MariaDB is a first-class engine, not an afterthought — the same driver, monitoring and editing you get with MySQL.
  • CockroachDB — Distributed SQL over the Postgres wire, with monitoring that reads CockroachDB’s own internal tables rather than pretending to be Postgres.
  • YugabyteDB — Distributed SQL that reuses the real PostgreSQL query layer — so AddisDB’s Postgres tooling, including the Live Monitor, works against it.
  • TiDB — MySQL-compatible distributed SQL with an HTAP twist. AddisDB drives it with the MySQL driver, monitoring included.
  • SingleStore — Real-time distributed SQL on the MySQL wire — rowstore and columnstore in one engine, with AddisDB’s full MySQL toolset on top.
  • Microsoft SQL Server — T-SQL with a native driver, DMV-backed live monitoring, and mock data that handles IDENTITY columns correctly.
  • Oracle Database — A native OCI connection to Oracle — no JDBC, no middleware. It needs Oracle Instant Client on your machine; here is how to get it.
  • PostGIS — Spatial Postgres, with geometry decoded automatically and plotted on the Map view — no ST_AsGeoJSON gymnastics required.
  • SQLite — Point AddisDB at a .db file and start querying. Full editing, mock data, and disposable test copies of the whole database.
  • DuckDB — In-process analytical SQL over a local file — and the engine quietly powering every CSV, Parquet and Excel file you open in AddisDB.
  • Greenplum — Massively-parallel analytics built on PostgreSQL. AddisDB connects with the Postgres driver, with introspection relaxed for Greenplum’s older catalog.
  • Aurora DSQL — AWS’s serverless distributed SQL, PostgreSQL-compatible on the wire. AddisDB connects with the Postgres driver; the catalog it exposes is deliberately small.
  • AlloyDB for PostgreSQL — Google’s PostgreSQL-compatible database with a columnar engine bolted on. Full Postgres catalog, so everything AddisDB does for Postgres works here.
  • Cloud SQL for PostgreSQL — Google-managed PostgreSQL. It is stock Postgres, so every AddisDB Postgres feature — Live Monitor included — works against it.
  • Cloud SQL for MySQL — Google-managed MySQL. AddisDB drives it with the MySQL driver, monitoring and mock data included.
  • Cloud SQL for SQL Server — Google-managed SQL Server. AddisDB drives it with the SQL Server driver, T-SQL batches and DMV monitoring intact.
  • PostgreSQL on Amazon RDS — Amazon-managed stock PostgreSQL. AddisDB connects with its Postgres driver and every Postgres feature in the app works, including the Live Monitor and test clones.
  • Aurora PostgreSQL — AWS Aurora with the PostgreSQL engine. Wire-compatible with Postgres and served by the same driver, so the full AddisDB toolset applies — including live monitoring.
  • Supabase — Supabase is a managed PostgreSQL with an API layer on top. AddisDB talks straight to the database with its Postgres driver — full catalog, full toolset.
  • Neon — Serverless PostgreSQL with branching. Stock catalog, so AddisDB’s Postgres driver gives you the full toolset — minus test clones, which Neon’s own branches replace.
  • EDB Postgres Advanced Server — EnterpriseDB’s PostgreSQL distribution with Oracle compatibility layered on. Stock catalog, so the full AddisDB Postgres toolset applies. Note the default port is 5444.
  • openGauss — A PostgreSQL fork from Huawei, and the base of the commercial GaussDB. Speaks the PG wire protocol; AddisDB relaxes introspection for its older catalog.
  • CrateDB — Distributed SQL for time-series and IoT that speaks the PostgreSQL wire protocol. AddisDB queries and browses it; the Postgres-specific tooling stays switched off.
  • Materialize — A streaming database that keeps SQL views incrementally up to date. Speaks the PG wire protocol on port 6875; AddisDB queries and browses it.
  • RisingWave — A streaming database with a PostgreSQL-compatible front end on port 4566. AddisDB queries and browses it; Postgres-specific tooling stays off.
  • MySQL on Amazon RDS — Amazon-managed stock MySQL. AddisDB connects with its MySQL driver and the full MySQL toolset applies, including the Live Monitor.
  • Aurora MySQL — AWS Aurora with the MySQL engine — wire-compatible with MySQL and served by the same driver, so the full MySQL toolset applies.
  • PlanetScale — Managed Vitess with a MySQL-compatible endpoint. AddisDB connects with its MySQL driver; branch-and-deploy-request workflow replaces direct schema edits.
  • Vitess — Horizontally-sharded MySQL. AddisDB connects to vtgate, which speaks the MySQL wire protocol, on port 15306.
  • Percona Server for MySQL — A drop-in MySQL distribution with extra instrumentation. AddisDB treats it exactly as MySQL — the full toolset applies.
  • OceanBase — A distributed SQL database. In MySQL mode it speaks the MySQL wire protocol on port 2881, and AddisDB connects with its MySQL driver.
  • Turso / libSQL — A local libSQL database file. libSQL forks SQLite and keeps its on-disk format, so AddisDB opens the file directly with its SQLite driver.
  • Azure SQL Database — Microsoft’s managed SQL Server, one logical database per connection. AddisDB connects with its SQL Server driver over TDS.
  • Azure SQL Managed Instance — Near-full SQL Server as a managed service, with an instance-wide catalog. AddisDB connects with its SQL Server driver over TDS.
  • Turso Cloud — The remote half of libSQL — SQLite reached over the Hrana v2 HTTP protocol. Local libSQL files have their own entry.
  • Cloudflare D1 — SQLite at Cloudflare’s edge, reached through the Cloudflare REST API. None of the connection fields is a hostname.
  • Firebird — The open-source RDBMS descended from InterBase. Needs a build with the firebird feature enabled; no native client library required.
  • SAP HANA — SAP’s in-memory column-store RDBMS. Needs a build with the hana feature enabled; the driver is pure Rust, so no SAP client library is required.
  • MongoDB — Run any Mongo command from the editor, get nested documents flattened into a grid, and keep the raw JSON one tab away.
  • Couchbase — Documents queried with N1QL (SQL++) — SQL you already know, over JSON. Buckets, scopes and collections all show up in the sidebar.
  • Azure Cosmos DB — Query the Core (SQL) API directly with your account key. AddisDB signs every request for you — no SDK, no emulator.
  • Firestore — Browse and query Google’s serverless document store with a gcloud access token. Read-only by construction — the driver has no write path.
  • Redis — Type Redis commands straight into the editor and read replies as a grid — arrays become rows, hashes become a keyed row.
  • Amazon DynamoDB — Query DynamoDB with PartiQL using your existing AWS credentials — nothing to paste, nothing for us to store.
  • Apache Cassandra — CQL over the native protocol, with keyspaces and tables — including partition and clustering keys — read straight from the system schema.
  • ScyllaDB — Cassandra-compatible and written in C++ for latency. Same CQL, same driver in AddisDB, same schema introspection.
  • Neo4j — Cypher over Bolt, with node labels in the sidebar and results flattened into a grid you can chart or read as JSON.
  • Memgraph — In-memory graph that speaks Bolt and Cypher, so it rides the same driver as Neo4j — with no password by default.
  • Elasticsearch — Type an index name, a raw query body, or a full REST call — hits come back flattened into the grid with _id and _score attached.
  • OpenSearch — The Apache-licensed Elasticsearch fork, with the same REST API — and the same three ways to query it in AddisDB.
  • Amazon DocumentDB — AWS-managed documents with MongoDB API compatibility. AddisDB connects with its Mongo driver — the same editor, the same grid.
  • Amazon Keyspaces — Serverless, Cassandra-compatible CQL from AWS. AddisDB connects with its Cassandra driver — on 9142, and TLS-only.
  • Apache CouchDB — A document store whose entire API is HTTP and JSON. Unrelated to Couchbase beyond the name.
  • Google Cloud Bigtable — Google’s wide-column store, reached over its REST API. None of the connection fields is a hostname.
  • pgvector — Embeddings in Postgres, plotted on the Vectors view — and every other Postgres feature AddisDB has, in the same connection.
  • Qdrant — A purpose-built vector database. Type a collection name to scroll its points, or write a raw search body — payloads land in the grid, embeddings on the Vectors view.
  • Weaviate — Query by class name or with GraphQL. Objects come back with their properties as columns and their embeddings on the Vectors view.
  • Pinecone — Point a connection at one index host and browse namespaces, metadata and scores — with the embeddings on the Vectors view.
  • Snowflake — The cloud warehouse, in the Big Data Console — browse the catalog, cap your previews, and estimate what a query will scan before you run it.
  • Google BigQuery — The one warehouse where cost estimation is exact — AddisDB runs a dry-run job and tells you the bytes before you are billed for them.
  • Databricks — Query the lakehouse over Delta Lake from the Big Data Console, with plan-based cost estimates and a catalog browser over Unity Catalog.
  • Amazon Redshift — AWS’s warehouse over the Postgres wire, opening in the Big Data Console with planner-based cost estimates.
  • ClickHouse — Columnar OLAP over the HTTP interface, with EXPLAIN ESTIMATE wired into the cost banner for MergeTree tables.
  • Trino — One SQL engine over every catalog you have connected. AddisDB follows Trino’s paged protocol and estimates IO before you run.
  • Presto — The engine Trino was forked from, served by the same AddisDB driver with Presto’s own header namespace.
  • Dremio — Lakehouse SQL over its REST job API — AddisDB submits, polls and fetches so a long job behaves like any other query.
  • Apache Hive — HiveServer2 over Thrift, driven straight from AddisDB — no JDBC driver, no Beeline. It needs HTTP transport enabled.
  • TimescaleDB — Time-series in real Postgres — so you get the full AddisDB toolset, Live Monitor and test clones included.
  • QuestDB — High-performance time-series SQL over the Postgres wire, on QuestDB’s own port 8812.
  • InfluxDB — Both query languages in one connection — InfluxQL by default, SQL with a toggle. AddisDB switches endpoints for you.
  • Prometheus — PromQL in a real editor, with a range-query directive that turns any expression into a chart.
  • Amazon Timestream — Serverless time-series on AWS, queried with SQL through your existing AWS credentials — nothing to paste, nothing stored.
  • StarRocks — A columnar OLAP database whose front end speaks the MySQL wire protocol. AddisDB connects with its MySQL driver on port 9030 and opens the Big Data Console.
  • Apache Doris — A columnar MPP analytical database with a MySQL-compatible front end on port 9030. AddisDB connects with its MySQL driver and opens the Big Data Console.
  • Databend — A cloud-native columnar warehouse with a MySQL-compatible handler on port 3307. AddisDB connects with its MySQL driver and opens the Big Data Console.
  • Azure Synapse Analytics — Synapse’s dedicated SQL pool is an MPP warehouse reached over TDS. AddisDB connects with its SQL Server driver and opens the Big Data Console.
  • Amazon Athena — Serverless SQL over data in S3. Reached through the AWS SDK with the ambient credential chain, so AddisDB stores no secret for it.
  • Firebolt — Cloud data warehouse over its HTTP query API, authenticated with a service-account access token.
  • MonetDB — An open-source column-store analytics database. AddisDB speaks its MAPI wire protocol directly.
  • Exasol — In-memory MPP analytics database, reached over its JSON-over-WebSocket protocol. Needs a build with the exasol feature enabled.
  • Apache Impala — MPP SQL over Hadoop and object storage. Speaks the same HiveServer2 protocol as Hive, on its own port.
  • Chroma — An open-source vector database for embeddings, reached over its REST v2 API.
  • Milvus — A vector database built for scale, reached over its REST v2 API — no gRPC tooling required.
  • CSV files — Open a CSV and query it with real SQL — joins, aggregates and all — without importing it into a database first.
  • JSON, NDJSON & JSONB files — Query a JSON export with SQL — arrays and nested objects flattened into columns, with the raw structure one tab away.
  • Excel (XLSX) files — Treat a spreadsheet as a table and ask it SQL questions — no formulas, no pivot tables, no manual re-sorting.
  • GeoJSON files — Open a GeoJSON file and its features land on an interactive map — with the properties queryable as a normal table.
  • Shapefiles (ESRI) — Pick the .shp and AddisDB finds its siblings, decodes the geometry, and plots it on the Map view — no GIS install required.
  • GeoPackage files — The modern OGC replacement for the shapefile — one self-contained file, opened and mapped the same way.
  • Feather files — The fast handoff format between Python and R sessions — open one directly and query it with SQL.
  • Apache Arrow files — Arrow IPC files opened directly, columnar types intact, queryable with SQL.
  • Parquet files — The format worth converting everything else into — compact, columnar, and fast to scan. Open one and query it in place.
  • DBF (dBASE) files — Read legacy dBASE tables — including the attribute file inside a shapefile set — with modern SQL.

Features & workflows

  • Find your slowest queries with the Profiler — See which statements actually cost your database time — from the engine’s own accounting, or from your own session when the engine can’t tell you.
  • Reading a query plan — Ask the engine how it intends to run a statement before you run it — and learn the handful of things worth looking for in the answer.
  • Read your schema as a diagram — The Schema tab lays your tables out as a clean, layered chart of foreign-key relationships — and lets you save named views per project.
  • Editing data in the grid — no SQL — Change a cell, add or delete a row, or drop a table right in the results grid — with a safety toggle so a stray click can never touch your data.
  • Compare two schemas with Diff — See exactly what differs between two databases — or your ORM schema and a live one — and get the SQL to bring them into line, one reviewed edit at a time.
  • Generating and applying migrations — Work with your Drizzle or Prisma migrations from inside AddisDB — generate the SQL, read every line, and apply it against the database you choose.
  • The Live Monitor and flight recorder — Watch your database in real time — active queries, blocking, waiting locks — and replay the moment something flared up, even after it is gone.
  • Automating work with workflows — Chain the things you keep doing by hand — queries, exports, backups, checks and drift reports — into one saved run.
  • Using the built-in terminal — A real shell docked under your work, with a consent step first — and a rule that it never types your password for you.
  • Backing up and restoring a database — Dump a whole database to a single file and load it back — using your engine’s own tool when it is installed.
  • Importing a CSV file — Load a delimited file into a table, with the format guessed for you and a preview before anything is written.
  • Exporting query results — Send a result set — or just the part you selected — to CSV, TSV, JSON, NDJSON or SQL.
  • Creating tables and databases — Build a table, database or schema through a form, review the SQL it produces, and change or drop objects safely.
  • Managing database users and privileges — See who can reach your database and what they are allowed to do, in a grid rather than a pile of GRANT statements.
  • Running one query across databases — Ask the same question of staging and prod at once and see exactly where the answers differ.
  • Applying a plain-SQL file — Open any .sql file, read a per-statement safety plan, rehearse it on a throwaway copy, then apply it.
  • Setting up alerts — Get told when a number crosses a line — from a metric AddisDB already watches, or a query you write.
  • Filtering results without SQL — Narrow a result set with a visual filter builder — including groups that mix AND with OR.
  • Saving and reusing queries — Star the queries you keep coming back to, and parameterise the ones you run with different values.

Safety & guardrails

  • How safe mode protects production — Read-only by default, destructive-statement detection, and a review step before anything touches prod.
  • Test your SQL on a throwaway clone — Dry-run a whole SQL file against a disposable copy of your database — data and all — so a risky migration never touches the real thing.
  • Find personal data with Data Map — Scan every connected database read-only and get an inventory of where personal, health and cardholder data actually lives.
  • Signet for Mac — A menu bar app that shows you exactly what an AI tool wants to do and waits for your fingerprint before it happens.

AI & keys

  • Bring your own AI key — AddisDB AI is free when you supply your own provider key. Here is where it is stored and how to rotate it.
  • Generating and removing mock data — Fill a table with realistic AI-generated rows, see them tinted amber, and remove them in one click when you are done.
  • Ask your database questions in plain English — The Chat tab turns a plain-English question into SQL against your real schema — and you review the query before it runs.
  • Keeping data away from AI — Lock tables and columns so nothing in AddisDB can send them to a model — any model, hosted or local.
  • Turning AI features off — Switch off every AI feature at once, one feature at a time, or only inside a single project — and see what AddisDB looks like while it is off.

Teams & billing

  • Teams, shared connections & billing — What a team plan unlocks — shared connection state, shared notebooks, and governance — and how billing works.
  • Sharing a connection securely — Hand a working connection to someone end-to-end encrypted, after verifying it is really them.
  • Team AI policy — Owners and admins on a Team or Business plan can turn AI off for everyone in their org, or feature by feature — and members cannot switch it back on.

Troubleshooting