AI Agent: Chat with Your PostgreSQL Database

Unlock Conversational Data Access with this AI Agent

This n8n workflow transforms how you interact with your PostgreSQL database. It's an AI Agent designed to understand your questions in plain English, convert them into precise SQL queries, and retrieve the information you need – all through a simple chat interface. No more complex SQL syntax for quick lookups; just ask, and the agent delivers.

This AI Agent acts as your personal 'DB assistant,' leveraging OpenAI's powerful language models (like GPT-4o mini by default) to interpret your requests. It's equipped with tools to understand your database structure, including fetching schema and table definitions, ensuring accurate query generation. It also maintains chat history for context-aware conversations, making follow-up questions intuitive.

Key Features & Benefits

• Natural Language Querying: Chat with your PostgreSQL database in plain English, no SQL knowledge required for basic queries.
• AI-Powered SQL Generation: Utilizes OpenAI's GPT models (e.g., GPT-4o mini by default) to translate your requests into executable SQL queries.
• Database Schema Awareness: Intelligently uses tools to fetch database schema, table lists, and table definitions to improve query accuracy.
• Contextual Conversations: Remembers previous interactions in the session for more natural follow-up questions using built-in chat memory.
• Democratizes Data Access: Enables non-technical team members (sales, support, marketing) to self-serve data insights.
• Streamlined Data Retrieval: Get answers from your database quickly without manual query writing or waiting for a data analyst.
• Customizable AI Behavior: Modify the system prompt in the AI Agent node to tailor the AI agent's responses and query generation strategy.

• An n8n instance (Cloud or self-hosted).
• OpenAI API Key with access to a suitable model (e.g., gpt-4o-mini or gpt-4).
• Credentials for your PostgreSQL database.

1. Download the n8n workflow JSON file.
2. Import the workflow into your n8n instance.
3. Set up your credentials:
   • OpenAI: In the 'OpenAI Chat Model' node, select or create your OpenAI credentials. Ensure a model like 'gpt-4o-mini' is chosen by default, or select another compatible model.
   • PostgreSQL: In the 'Execute SQL Query', 'Get DB Schema and Tables List', and 'Get Table Definition' nodes, select or create your PostgreSQL credentials.
4. (Optional) Customize the systemMessage in the 'AI Agent' node to fine-tune the AI's behavior and its understanding of your database. The default prompt guides the AI to use its tools (schema fetching, query execution) effectively.
5. (Optional) Adjust the 'Chat History' node settings, like Context Window Length, to control how many past messages are kept for context (default is 5 messages).
6. Activate the workflow.
7. To start chatting, click the 'Chat' button on the 'When chat message received' node within the n8n editor. For production use, you can use the production webhook URL if you make the chat publicly available.