AIToday

Dribble: Open-source AI SQL IDE for database queries

Hacker News6h ago4 min read
Dribble: Open-source AI SQL IDE for database queries

Key takeaway

Dribble is a new open-source SQL IDE that pairs a traditional query editor with an embedded AI agent, allowing users to write SQL and ask natural-language questions about their database schema in a single tabbed workspace. The tool persists queries, results, and chat history server-side, supports multi-user deployments with optional Google authentication, and encrypts stored database credentials. It currently supports Postgres and is built to add drivers for other databases.

Summaries like this, in your inbox every morning.

Sign up free →

3 Key Points

  • What happened

    Dribble, an open-source web-based SQL IDE, launched with built-in AI data analyst capabilities. Users can connect to Postgres databases, browse schemas, run queries in a notebook interface, and chat with an AI agent (Claude Opus 4.8) that inspects schemas, writes read-only SQL, and iterates on errors.

  • Why it matters

    The tool combines SQL development and AI-assisted data exploration in one workspace, potentially reducing the time developers and analysts spend switching between query editors and chat interfaces. State is stored server-side and persists across browser sessions, and the project supports multi-user deployments with optional Google sign-in and encrypted credential storage.

  • What to watch

    Dribble is released under the MIT License with no login required for local use. The driver registry is designed to add support for additional databases (MySQL, Snowflake, and others) beyond the current Postgres support. The codebase was written largely with AI coding tools and has been reviewed before commit, though the maintainers recommend review before production use.

FAQ

What databases does Dribble support?
Postgres ships with Dribble today. The driver registry is built to add more engines such as MySQL and Snowflake.
Do I need to log in to use Dribble?
No login is required for local use. For multi-user deployments, you can optionally configure Google sign-in with an email or domain allowlist to keep each person's connections, notebooks, and chats private.
What AI model powers the assistant?
The AI data analyst uses Claude Opus 4.8 and requires an Anthropic API key to run.

Discussion

No comments yet. Be the first to share your thoughts!

Log in to join the discussion

Related Articles

Stay ahead with AI news

Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.

Get Started Free

Free · takes 30 seconds · unsubscribe anytime

1 minute a day. The AI essentials.

200+ sources · Email / LINE / Slack

Get it free →