Data & Analytics

Wren AI

by Canner, Inc

Overview

Wren AI brings conversational business intelligence (GenBI) into ChatGPT. Connect your company's databases, ask questions in free-form language, and get answers from live data with the SQL, visuals, and context behind them. Ask "What was revenue by region last quarter?" and Wren AI generates the SQL, runs it against your warehouse, and returns the result with a short summary or chart. Every query is built on your team's semantic model - the business definitions, metrics, and joins your data team already maintains - so ChatGPT uses the same terms and logic your analysts do instead of guessing at table names. What you can do: - Ask questions and get answers with the SQL shown - Generate SQL from a question without running it - Run a query and get results back - Summarize results or turn them into a chart - Switch between the Wren AI projects you have access to

Tools

ask

ChatGPT

This MCP server exposes multiple projects behind one endpoint. First call list_projects to choose a projectId, then call get_project_metadata with that projectId to understand the available tables and columns before asking. Pass the same projectId to every tool call. Ask a natural language question about the selected project's data (via projectId) and receive a comprehensive answer including SQL query, executed data results, and summary. Supports conversation threading for context-aware follow-up questions.

generate_chart

ChatGPT

Generate a chart from a question and SQL query for the selected project (projectId from list_projects).

generate_sql

ChatGPT

This MCP server exposes multiple projects behind one endpoint. First call list_projects to choose a projectId, then call get_project_metadata with that projectId to understand the available tables and columns before asking. Pass the same projectId to every tool call. Generate SQL queries from natural language questions for the selected project (via projectId), without generating summaries.

generate_summary

ChatGPT

Execute a SQL query against the selected project (projectId from list_projects) and generate a natural language summary of the results. Useful when you already have SQL and want insights without regenerating it.

get_project_metadata

ChatGPT

Get the MDL for a specific project. Requires a projectId from list_projects. MDL (Modeling Definition Language) is a declarative language used to describe tables, fields, relationships, metrics and business rules in a structured, code-like format. It turns raw database schemas into a semantic model that LLMs and tools can understand and use safely. Call this before ask/generate_sql to learn the project's tables and columns.

list_projects

ChatGPT

List the projects you are allowed to query. Use the returned id as the projectId argument for the other tools.

respond_clarification

ChatGPT

When the ask tool returns a "needs_clarification" result, present the question to the user, then call this tool with their answer to resume the ask and get the final result. Required fields — queryId, questionId, threadId — come from that result. Pass the chosen option value(s) in answers, and the same projectId you used for ask.

run_sql

ChatGPT

Execute a read-only SQL query against the selected project's warehouse (projectId from list_projects) and receive raw data results without any AI-generated summaries or transformations.

ask

Claude

generate_chart

Claude

generate_sql

Claude

generate_summary

Claude

get_project_metadata

Claude

list_projects

Claude

respond_clarification

Claude

run_sql

Claude

Capabilities

Interactive

Example Prompts

Click any prompt to copy it.

Links

Documentation ↗Privacy Policy ↗

App Stats

Tools

Prompts

ChatGPT, Claude

Platforms