Turn plain questions into live database answers. A chat message triggers an AI agent that reads your Supabase tables, writes safe SQL, and returns summaries and insights. Great for teams that need quick data checks without writing SQL.

Incoming chat text flows into an AI agent powered by an OpenAI model. The agent uses three database tools. One lists all public tables. One returns a full column map for any table, including types and references. One runs the final SQL. Results come back in natural language with counts, trends, and JSON field parsing when needed. The design reduces back and forth with analysts and keeps the conversation in one place.

You need OpenAI and a Supabase Postgres database. Add your Supabase host, database, user, and password in the Postgres nodes. Set the system message to guide the assistant style. Expect faster ad hoc reporting, fewer query errors, and quicker decisions for operations, product, and support analytics.

• Chat trigger captures each user message and starts the analysis flow
• OpenAI chat model interprets intent and drafts SQL in natural language
• Database schema tool lists all public tables to guide correct table selection
• Table definition tool returns columns, types, and references for precise queries
• Run SQL Query tool executes final statements and returns rows for analysis
• Agent combines tools and language model to summarize results and explain trends
• System prompt can be fine tuned to match your data rules and tone

• Reduce manual SQL writing from hours to minutes by turning questions into automatic queries
• Streamline ad hoc reporting by up to 70 percent with a single chat entry point
• Improve data accuracy by cutting query syntax mistakes by 50 percent using schema aware tools
• Handle more requests without extra staff by letting non technical users self serve
• Connect OpenAI and Supabase in one flow for faster insight delivery

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with OpenAI and Supabase. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, create a new OpenAI credential. Generate an API key in your OpenAI account, then paste it into the API Key field and save.
4. In Supabase, open Project Settings and copy the database host, port, database name, user, and password. Make sure the role has read access to the needed schemas.
5. In the n8n credentials manager, create a new PostgreSQL credential using your Supabase details. Test the connection and save.
6. Open the OpenAI Chat Model node. In the credential dropdown, choose your OpenAI credential. Pick the desired model and save.
7. Open the DB Schema, Get table definition, and Run SQL Query nodes. In each, select the PostgreSQL credential you created. Keep the provided SQL queries as is.
8. Open the AI Agent node. Confirm the language model is set to the OpenAI Chat Model node and that all three database tools are attached as agent tools.
9. Adjust the system message in the AI Agent node to match your data policies and answer style. You can require safe aggregation or limits on row counts.
10. Click Execute Workflow and send a test message in the chat trigger. Ask for a simple count from a known table to confirm access.
11. If queries return empty results, check table names in the DB Schema tool output. If you see permission errors, update Supabase role grants for the public schema.
12. For SQL errors, open the Table definition tool output to verify column names and data types, then retry the question with clearer terms.