System Prompt Generator

Answer a few questions about identity, knowledge, tone, and rules to assemble a clear system prompt for building an AI assistant, chatbot, or custom GPT.

Identity / roleKnowledge / domainTone and styleRules (one per line)Response formatOut-of-scope handling

System prompt

You are a customer support assistant for a SaaS project management tool.

Your knowledge covers: Our product's features, pricing plans, and common troubleshooting steps.

Tone and style: Friendly, patient, and concise.

Rules you must follow:
- Never make up pricing or features you're not sure about.
- Ask a clarifying question if the request is ambiguous.
- Escalate billing disputes to a human agent instead of resolving them yourself.

Response format: Plain, conversational text. Use short paragraphs, not bullet lists, unless listing steps.

Out-of-scope requests: If asked something unrelated to the product, politely redirect the conversation back to how you can help.

Ad space

Frequently asked questions

What's a system prompt used for?+

It's the instruction set given to an AI model before any user messages, defining its role, knowledge boundaries, tone, and rules for an entire conversation or application.

How is this different from the AI Prompt Generator?+

The AI Prompt Generator builds a one-off task prompt; this tool builds a persistent system-level prompt meant to govern an AI's behavior across an entire session or app.

Where do I use the generated prompt?+

In the "system" role field of an API call (OpenAI, Anthropic, and most providers support this), or in the custom instructions field of tools like custom GPTs.