DecisionBox

Autonomous AI discovery on your data — open source

166 followers

Autonomous AI discovery on your data — open source

166 followers

•

•

DecisionBox is an autonomous AI agent that connects to your data warehouse, writes its own SQL, validates every finding against your data, and ships a ranked backlog of insights and recommendations. No prompting. No question-writing. Open source, AGPL v3.

This is the 10th launch from DecisionBox. View more

DecisionBox for Microsoft SQL Server

Launched this week

Connect DecisionBox to your MSSQL to validate findings

Connect DecisionBox to your Microsoft SQL Server — on-prem, in the cloud, or Azure SQL. The agent writes its own SQL, validates every finding against your data, and ships a ranked backlog — no prompting. Read-only by SQL Server logins, encrypted by default. Open source, AGPL v3.

DecisionBox for Microsoft SQL Server gallery image

Free Options

Launch tags:Open Source•Developer Tools•Artificial Intelligence

Launch Team

FramerLaunch websites with enterprise needs at startup speeds.

Promoted

DecisionBox

Maker

📌

Hey PH, Selçuk here from DecisionBox, If you're on SQL Server or Azure SQL, the database that runs your app already holds the data worth analyzing. The hard part isn't getting to it. It's figuring out which questions are worth asking, getting validated answers, and doing it without disturbing the workload the database is already serving. That's what we built DecisionBox for. Point it at a SQL Server, a database, and a schema. The agent writes its own SQL, validates every finding against the data, and ships a ranked backlog of insights and recommendations. No prompting, no question-writing. Today we're adding Microsoft SQL Server to the list of supported warehouses — and unlike most things in this category, the provider is built for a database that's also serving your app. Read-only, scoped by a SQL Server login. Create a login (or a contained user on Azure SQL) with SELECT on the schemas and tables you want exposed, nothing more. The GRANT model is the boundary; the agent cannot reach anything it hasn't been granted. Safe to run on a production database. Small connection pool, capped at 5. Per-query timeout, defaults to 5 minutes. Row counts read from sys.dm_db_partition_stats — engine-maintained, no full table scan to count rows. TDS encryption on by default. No writes, ever — discovery queries are read-only SELECTs. Two ways in, both standard SQL Server: a username/password login, or a full sqlserver:// connection string for Azure SQL or anything that needs extra TDS parameters. If your security team has already approved how the rest of your stack talks to this database, they have already approved DecisionBox. The whole SQL Server provider is in the public repo, AGPL v3 — the auth flow, the INFORMATION_SCHEMA reads, the T-SQL the agent writes. Read it before you turn it on. Same agent already runs against BigQuery, Redshift, Snowflake, Postgres, and Databricks. If your stack moves, your DecisionBox install moves with it. Happy to dig into the SQL login setup, Azure SQL specifics, or anything else in the comments.

Report

3d ago

mailX by mailwarm

Congrats on your launch. Wishing you the best!

Report

3d ago

DecisionBox

Maker

@daniel_nwankwo Thank you so much, Daniel! 🙏

Report

3d ago

Previous DecisionBox Launches

DecisionBox for DatabricksConnect DecisionBox to your Databricks to validate findings

Launched on May 22nd, 2026

DecisionBox for PostgreSQLAutonomous AI discovery, safe on a production Postgres.

Launched on May 11th, 2026

DecisionBox for SnowflakeAutonomous AI discovery on Snowflake

Launched on May 8th, 2026

DecisionBox for Amazon RedshiftAutonomous AI discovery on Amazon Redshift.

Launched on May 4th, 2026

View all DecisionBox launches

Forum Threads

p/decisionbox

•

1mo ago

Domain Packs: teach DecisionBox your industry in minutes

Domain Packs: Teach DecisionBox Your Industry in Minutes

Create, edit, import, and share AI discovery templates entirely from the dashboard.

Cross-posting from our blog for anyone who hasn't seen it. Original post: https://decisionbox.io/blog/doma...

TL;DR

p/decisionbox

•

2mo ago

We pointed DecisionBox at a data set it had never seen. Here's what it found under an hour.

Most AI analytics tools need you to ask the right questions. We wanted to see what happens when you don't ask anything at all.

We took a public Kaggle dataset, a 109 million raw e-commerce dataset from a multi-category store. No documentation, no data dictionary, no context. We pointed DecisionBox at it and hit "Start."

p/decisionbox

•

2mo ago

DecisionBox AI Agent ran 233 SQL queries on Databricks in 76 minutes. Here's what it found.

Hey Product Hunt!

I'm Can, co-founder of DecisionBox. I wanted to share a real case study of what our open-source AI agent does when you point it at a data warehouse.

View all