wrenai-analyst

You can generate SQL from natural language by asking your data question directly, which triggers the system to translate it into executable SQL, run the query, and return results. It reads your connected schema automatically to understand the data model.

Yes, you can visualize SQL query results as charts by simply asking to chart or visualize the data. The system generates Vega charts directly from your natural language request and the executed query results.

Yes, you need a connected WrenAI MCP to run natural language data queries. You must set it up first using the /setup-wrenai-mcp command to load project metadata and enable SQL generation, execution, and charting.

The best way to get actionable business insights from ad hoc data analysis is to ask natural language questions against your data model. The system executes queries and proactively generates summaries and recommendations based on the results.

Schema auto-discovery works by performing a get_project_metadata operation when you ask a question. It reads the connected data model structure automatically, so you do not need to manually explain table relationships or schema details before querying.

Natural language to SQL generation for complex BI queries depends on the connected data model's structure and the WrenAI MCP's schema understanding. It does not support querying disconnected data sources or generating visuals outside the supported Vega chart format.