Verified project record
Canner/WrenAI
An open-source GenBI engine that lets AI agents generate, deploy, and govern business intelligence using a version-controlled context layer across multiple data sources. It is designed for teams who need governed SQL and dashboard generation driven by business semantics.
Project overview
It provides a reviewable, Git-friendly context layer (MDL, instructions.md) that captures business semantics and governance, aiming to make AI-generated BI outputs more trustworthy.
- Project type
- RAG · Data Processing · Prompt Engineering
- Use cases
- Knowledge Q&A · Data Analysis
- Deployment
- Refer to project documentation
- License
- License pending
Best for
- Developers, data teams, and enterprise teams who need to generate, deploy, and govern business intelligence using AI agents.
- Users who manage multiple data sources and require a version-controlled context layer to maintain business semantics.
Key capabilities
- Turns natural-language business questions into governed SQL and charts using schema-aware retrieval, MDL planning, and dry-plan validation.
- Provides a reviewable, Git-friendly semantic layer capturing business semantics, definitions, memory, and governance.
- Builds interactive browser-side dashboards and deploys them to hosting platforms like Vercel or Cloudflare Pages.
- Provides reference agent SDK integrations for LangChain/LangGraph and Python, enabling existing agents to drive the BI workflows.
- Provides hybrid retrieval for recalling similar past queries using a local LanceDB memory index.
- Includes functions, dry-plans, row limits, and row-level/column-level access control for secure query execution.
Limitations and risks
- The earlier Docker-based chat-first BI product (Wren AI GenBI) is on a legacy branch with no new features or security fixes.
- Paid services may be used optionally, such as hosting platforms for dashboard deployment.
- Telemetry practices are not documented. Minimum hardware requirements are not specified.
Getting started
- Setup is documented as easy because it is agent-driven by design: install the CLI, install a one-file stub, let the agent set up the database, and ask the agent a question. Coding requirements are optional.
- Users can start by running 'pip install wrenai', followed by 'npx skills add Canner/WrenAI' to install the required stub.
Alternatives and comparisons
- A Python library for asking questions to data (SQL, CSV, parquet) in natural language and generating visualizations.
- A Python Text-to-SQL RAG Library that streamlines interaction between users and databases.
- A PyTorch implementation using MCTS and self-rewarding to dynamically generate SQL queries.
Project comparisons
Evidence and sources
- Release: wren: v0.13.0
- GitHub project description: 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…
- GitHub repository metadata: {"primary_language":"Python"}
- README: WrenAI is the **open-source GenBI engine**: it lets AI agents **generate, deploy, and govern** business intelligence, from a SQL answer to a shareable dashboard, across 22+ data s…
- README: - **Generate.** Your agent turns a business question into *governed* SQL and charts. Schema-aware retrieval, MDL planning, dry-plan validation, and structured errors keep it corre…
AI Search
Find projects, verify facts, compare options, or turn a complex need into an actionable plan
Try a searchA click only fills the search box; you stay in control
Project Details
0