Wren AI

Wren AI is an open-source data agent that turns plain-language questions into trusted answers. Your team types a question in English, and Wren writes the SQL and returns the result. It learns your business through a context layer of metrics, relationships, and definitions that your data team approves. Every answer respects your access rules, so people only see data they are allowed to see. Claude, ChatGPT, and your own agents can ask it over MCP. It connects to Snowflake, BigQuery, PostgreSQL, and 20-plus more sources.

Key Features

Ask from Anywhere

Your team asks inside Wren, Slack, or Teams. Everyone gets the same answer, no matter where the question starts.

Shows Its Work

Every answer comes with its reasoning and the SQL it ran. You can audit exactly how Wren reached each result.

Data Stays in Place

Wren connects without ETL or migration. Your data stays where it is while you query it live.

Drillable Dashboards

Wren turns each question into a dashboard. You can drill into results and filter them without writing new SQL.

Embed in Your Product

You can put the same data agent inside your own app, under your brand, with Wren's commercial plan.

Run It Your Way

Deploy Wren on your own servers, in a private cloud, or fully air-gapped. Your data never has to leave your network.

FAQ

How does Wren learn my business?

The onboarding agent drafts your context layer from your warehouse. Your team corrects any mistakes before it goes live.

Does Wren AI replace my data team?

No. It handles routine questions so analysts spend less time on repeat ad-hoc requests and more on deep work.

What does "one trusted answer" mean?

Everyone sees the same answer to the same question. People and AI agents share one definition, so numbers never disagree.

Is Wren AI really open-source?

Yes, the core is open-source on GitHub with over 17,000 stars. You can inspect the code, run it yourself, and contribute.

Conclusion

Wren AI gives your whole team one trusted way to ask data questions. Plain-language answers, visible SQL, and rules your data team controls make it safe for non-technical users. AI agents get the same answers over MCP. If your analysts drown in ad-hoc requests, Wren can clear the queue. Connect a source and ask your first question.