n8nSoftwareShipping codeAI score · 75/100AI-assisted edit · no obvious prompt-injection riskn8n.io

Build a System Prompt Optimizer Agent with n8n + OpenAI

Input a draft prompt → AI Agent rewrites it into a polished system prompt and recommends a model → iterate with memory

Some experienceStepsWorkflowPrompt
Who
Developers or automation builders who want to generate or refine AI system prompts inside n8n
You get
A clear, actionable optimized system prompt plus an OpenAI model recommendation for the task
You need
n8n

Constraints

Prerequisites: n8n instance (self-hosted or cloud), OpenAI API key with available balance. The template has 8 nodes (4 functional + 4 sticky notes). Core flow: Chat Trigger → AI Agent → OpenAI Chat Model + Buffer Window Memory. Default model is gpt-5.

What they did

Author rbreen published an n8n template that wires Chat Trigger, AI Agent (LangChain), OpenAI Chat Model (gpt-5), and Buffer Window Memory into a System Prompt Optimizer Agent.

How it went

Template status is published with 2857 views (author_report / API data). No runtime logs or effect comparisons are provided by the source.

Worth knowing

The model field shows gpt-5; actual availability depends on OpenAI account permissions and API version. Memory is a simple Buffer Window and may drop older context in long conversations. Cost is not estimated by the source; billed per OpenAI pricing.

Steps

  1. Set up OpenAI credentials

    Create an OpenAI API credential in n8n. Go to OpenAI Platform to generate an API key, ensure the account has a usable balance, then paste the key into n8n’s OpenAI credentials.

  2. Import and verify workflow structure

    After importing the template, confirm it contains: Chat Trigger (input), AI Prompt Generator (Agent node with systemMessage), OpenAI Chat Model10 (model gpt-5), Simple Memory2 (Buffer Window Memory). Connections: Chat Trigger → AI Prompt Generator; OpenAI Chat Model → AI Prompt Generator (language model); Simple Memory → AI Prompt Generator (memory).

  3. Run and iterate

    Activate the workflow, then send a draft prompt through the Chat Trigger. The Agent returns an optimized system prompt and a model recommendation. The Memory node retains context so you can send follow-up refinements across multiple turns.

    Execution prompt · Source text。Chat Trigger input example (sticky-notes.md Sticky Note36 / description.md Example Question)

    I need a system prompt for an agent that summarizes meeting transcripts into 3 bullet points for executives.

Prompt you can copy

Execution prompt · Source text。Send it in the n8n chat window (Chat Trigger)

This block is the prompt. Ads are never placed here.

I need a system prompt for an agent that summarizes meeting transcripts into 3 bullet points for executives.

No standalone main prompt is required; the core execution prompt is embedded in the AI Prompt Generator node’s systemMessage.

Source

AI system prompt generator & optimizer (n8n + OpenAI) · rbreen · n8n.io · 2025-08-19T19:20:58.475Z

Comments & feedback

Tried this method? Share results, problems or improvements. Comments appear after moderation. Please avoid sensitive information.

Loading…

    Related