Wren AI

Overview

Wren AI is an open-source Agentic Generative BI (GenBI) platform designed for teams and AI agents that need reliable, governed access to enterprise data. By allowing users to ask questions in plain English, Wren AI returns governed SQL queries, interactive charts, and GenBI applications across more than 20 data sources. It is available as a fully managed cloud service or as a self-hosted deployment, giving organizations control over their data while leveraging an AI-native analytics layer.

At the core of Wren AI is a context engine that automatically constructs a semantic understanding of your data using MDL (Metric Definition Language). This semantic layer serves as the foundation for every answer, ensuring consistency across human and agent interactions. The platform enforces row-level policies at query execution time, so users and AI agents only see data they are permitted to access. The engine is open source and hosted on GitHub, while your semantic definitions can be version-controlled in your own git repository, creating a transparent and auditable analytics workflow.

Wren AI supports agentic workflows with a sandboxed, multi-step reasoning environment. AI agents can query data, generate charts, extract information from PDFs, build dashboards, and save reusable skills, with every step fully traceable and replayable. The same context, policies, and semantics are applied whether a human asks through natural language or an AI agent invokes a tool such as wren.query . This guarantees that answers remain consistent and governed across all modes of access.

For business teams, Wren AI transforms a single natural-language prompt into a live dashboard with built-in drill-down, roll-up, and global filters, enabling leaders and analysts to act on real-time data. With 20+ connectors, an embedded API, and open-core flexibility, Wren AI Cloud is suited for data teams, business leaders, and AI engineers seeking a trustworthy, governed GenBI layer across their organization.

Core capabilities

Text-to-SQL

Ask questions in plain English and receive accurate, generated SQL queries based on your semantic context.

Governed at execution

Security and access policies are enforced directly at query runtime, ensuring consistent compliance.

MDL semantics

A Metric Definition Language-based semantic layer automatically builds context for accurate, consistent answers.

20+ connectors

Natively connect to a wide range of data sources, in Wren AI Cloud or your own infrastructure.

GenBI apps

Generate interactive dashboards with drill-down, roll-up, and global filter capabilities from a single prompt.

Embedded API

Integrate governed GenBI features and agent query capabilities directly into your own products and workflows.

Agent skills & memory

AI agents can save reusable skills and maintain memory while executing sandboxed, multi-step reasoning.

Row-level policy

Enforce granular data access rules that apply uniformly to both human users and AI agents.

Open core

The core context engine is open source on GitHub, with definitions stored in your own git for transparency.

Cloud or self-hosted

Deploy as a managed cloud service or run it entirely within your own environment for full control.

Use cases

Natural-language querying of enterprise data with full governance and row-level security

Building real-time executive dashboards from a single prompt, such as expansion plans or revenue analysis

Enabling AI agents to perform governed text-to-SQL queries and structured data extraction

Self-service analytics across 20+ data sources without manually writing SQL

Embedding governed GenBI apps and agent APIs into internal tools or customer-facing products

Creating a shared semantic context so humans and AI agents stay aligned on metrics and definitions