WrenAI
GenBI (Generative BI) for AI agents, an open-source, governed text-to-SQL through an open context layer that turns natural-language questions into trusted dashboards, charts, and SQL across 20+ data sources, such as BigQuery, Snowflake, PostgreSQL, ClickHouse, Amazon Redshift, Databricks and more.
WrenAI is an open-source generative business intelligence (GenBI) engine that enables AI agents to transform natural-language questions into governed SQL queries, interactive dashboards, and shareable analytics across more than 22 data sources including BigQuery, Snowflake, PostgreSQL, ClickHouse, Amazon Redshift, and Databricks. Built on an AI context layer and semantic layer expressed in Modeling Definition Language (MDL), it provides trustworthy, version-controlled business definitions that agents can reliably use instead of guessing at SQL. Agents install lightweight discovery stubs for clients like Claude Code, Cursor, and Cline, then fetch workflow guides on-demand to set up databases, enrich context, generate queries, and deploy browser-based dashboards to Vercel or Cloudflare Pages.
Key Features
Use Cases
- 01Enable AI agents to answer business questions with trustworthy, governed SQL instead of hallucinated queries
- 02Automatically generate and deploy interactive dashboards from natural-language questions without manual BI tool configuration
- 03Maintain reviewable, version-controlled business definitions and data semantics that every agent and team member can use
- 04Build end-to-end generative BI workflows where agents onboard databases, enrich context, query data, and share results
- 05Integrate governed text-to-SQL capabilities into existing agent stacks using Claude Code, Cursor, Cline, or custom frameworks
- 06Query multiple data warehouses (BigQuery, Snowflake, Redshift, Databricks) through a unified semantic layer with consistent business logic
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
WrenAI — FAQ
What is WrenAI and what does it do?+
WrenAI is an open-source generative BI engine that lets AI agents convert natural-language questions into governed SQL queries, interactive dashboards, and shareable analytics across 22+ data sources. It uses a semantic layer (MDL) and AI context layer to ensure queries reflect approved business definitions instead of schema guesses.
How do I install the WrenAI agent skill?+
First install the Python CLI with 'pip install wrenai' (or with data-source extras like 'pip install wrenai[postgres,memory]'). Then add the discovery stub to your AI client by running 'npx skills add Canner/WrenAI', which auto-detects Claude Code, Cursor, Cline, and similar clients. Your agent can then fetch workflow guides on demand.
Which AI clients work with WrenAI?+
WrenAI works with Claude Code, Cursor, Cline, Codex, and other AI clients that support skill discovery. The 'npx skills add' command auto-detects your client and installs a lightweight stub that teaches your agent to fetch workflow guides and invoke the CLI.
Do I need API keys or a database connection?+
You need connection credentials for the data source you want to query (BigQuery, Snowflake, PostgreSQL, etc.). The agent uses 'wren skills get onboarding' to guide you through creating a connection profile. A bundled 'jaffle_shop' sample dataset is available for testing without your own database.
Is WrenAI free to use?+
Yes, the core engine—MDL semantic layer, governed text-to-SQL, CLI, MCP server, and 22+ connectors—is open source under Apache 2.0 and free forever. Row/column-level security, user access control, the GenBI UI, and enterprise features are commercial offerings available as Wren AI Cloud or self-hosted Enterprise Plus.
What data sources does WrenAI support?+
WrenAI supports 22+ data sources via an Apache DataFusion engine, including BigQuery, Snowflake, PostgreSQL, ClickHouse, Amazon Redshift, Databricks, DuckDB, MySQL, SQL Server, and more. Each source can be added as a pip extra or configured through the CLI.
How do I install WrenAI?+
Open the source repository on GitHub and follow its README. WrenAI is a skill — MCP Agents Market links you directly to the official repo.
Is WrenAI free?+
WrenAI is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.