Split test different agent prompts with Supabase and OpenAI
Split Test Agent Prompts with Supabase and OpenAI Use Case Oftentimes, it's useful to test different settings for a large language model in production against various metrics. Split testing is a good method for doing this. What it Does This workflow randomly assigns chat sessions to one of two prompts, the baseline and the alternative. The agent will use the same prompt for all interactions in that chat session. How it Works 1. When messages arrive, a table containing information regarding sessi
At a glance
Split test different agent prompts with Supabase and OpenAI is a ready-made n8n workflow you import as a workflow JSON file — no build required. It connects OpenAI. It's free to download. Follow the 5-step import below to go live in minutes.
- Platform
- n8n
- Connects
- OpenAI
- Modules
- 9
- Price
- Free
- Version
- v1.0

About this workflow
Split Test Agent Prompts with Supabase and OpenAI Use Case Oftentimes, it's useful to test different settings for a large language model in production against various metrics. Split testing is a good method for doing this. What it Does This workflow randomly assigns chat sessions to one of two prompts, the baseline and the alternative. The agent will use the same prompt for all interactions in that chat session. How it Works 1. When messages arrive, a table containing information regarding session ID and which prompt to use is checked to see if the chat already exists 2. If it does not, the session ID is added to the table and a prompt is randomly assigned 3. These values are then used to generate a response Setup 1. Create a table in Supabase called splittestsessions. It needs to have the following columns: sessionid (text) and showalternative (bool) 2. Add your Supabase, OpenAI, and PostgreSQL credentials 3. Modify the Define Path Values node to set the baseline and alternative prompt values. 4. Activate the workflow and test by sending messages through n8n's inbuilt chat 5. Experiment with different chat sessions to test see both prompts in action Next Steps - Modify the workflow to test different LLM settings such as temperature - Add a method to measure the efficacy of the two alternative prompts
How to import this n8n workflow
- 1
Download the workflow JSON file after purchase.
- 2
Open n8n → click the menu → Import from File.
- 3
Select the downloaded JSON and import.
- 4
Set up credentials for each node that requires them.
- 5
Click Execute Workflow to test, then activate.
Setup guide
Setup guide included
Purchase to unlock the full step-by-step guide
Related N8n workflows
AI image generator from text built on fal.ai
Free AI image generator - n8n automation workflow with Gemini/ChatGPT
Send to data to Google Sheets with ChatGPT using converted HTML to text from HTTP
Automatic Gmail categorization and labeling with AI
Ai-powered degree audit system with Google Sheets and GPT-5
Screen and score investment deals with AI using OpenAI, Gmail, and Telegram
Create and publish AI recipe infographics with Gemini, Nanobanana Pro and Blotato
Generate & publish professional COE blogs with Gemini AI and Google Drive
Reviews
No reviews yet
Be the first to buy and share your experience.
Leave a review
Sign in to share your experience with this workflow.
Create a free account to purchase workflows.
- JSON blueprint — instant download
- Setup guide PDF included
- 5 downloads · valid 30 days
- Works with n8n
Need help setting this up?
Book a 3-hour live setup session with an Agility consultant.
- Configure live on Google Meet / Zoom
- Free follow-up if workflow has defects
- Platform expert assigned to you