Wren AI

Wren AI is presented as an open-source GenBI (generative business intelligence) agent. People ask business questions in plain English, and AI agents such as Claude and ChatGPT can query the same data over MCP, with the goal that everyone receives the same answer based on definitions the organization controls. The page says Wren connects to existing data sources without ETL or migration, uses an onboarding agent to draft a context layer (metrics, relationships and definitions) that a data team reviews and approves, and returns answers that show their reasoning and the SQL executed. It also describes generating live, drillable dashboards from plain-English prompts, answering questions in Wren, Slack or Microsoft Teams, and offering Wren AI Cloud, VPC or fully on-premise deployment. The page lists 17K+ GitHub stars, SOC 2, and customers in regulated industries such as manufacturing, banking and healthcare.

Key features

• Natural-language questions over connected business data
• Shared context layer of metrics, relationships and business definitions, reviewed and approved by a data team
• Onboarding agent that drafts the context layer
• Access controls so users and agents only see permitted rows and columns
• Answers that show reasoning and the SQL that was run
• Live dashboards generated from plain-English prompts, with drill-down, roll-up and global filters
• MCP server so Claude, ChatGPT, Gemini and custom agents can query the same answers
• Access from Wren, Slack and Microsoft Teams
• Connectivity to data sources including Snowflake, BigQuery, Databricks, Redshift, PostgreSQL, SQL Server, Oracle, MySQL, ClickHouse, Athena, Trino and Starburst, with 20+ sources mentioned
• Embedding the data agent inside a customer's product
• Cloud, self-hosted VPC, or fully on-premise deployment

Best suited for

• Business teams that need trusted answers to data questions without writing SQL
• Data teams that want to govern metric definitions for both people and AI agents
• Organizations in regulated industries such as manufacturing, banking and healthcare
• Companies connecting AI assistants like Claude or ChatGPT to governed company data
• Software vendors embedding a data agent into their own products

Pricing

Not stated on the website