System prompts: how to set persistent instructions for ChatGPT, Claude, and Gemini

How to write system prompts (custom instructions) that shape every response without repeating yourself — covering persona, constraints, output format, and escalation rules.

← Back to guides

What system prompts are and when to use them

A system prompt is persistent context that runs before every user message — it shapes tone, persona, constraints, and behavior across the whole conversation.
Use system prompts when you need consistent behavior across many interactions: a support bot, a writing assistant, a code reviewer, an internal tool.
Without a system prompt, each request starts from scratch — you lose consistent tone, format, and safety rules between turns.
System prompts are available natively in ChatGPT (Custom Instructions or GPT system prompt), Claude (system parameter), and Gemini (system instruction).

What to put in a system prompt

Persona: who the model is, what domain it knows, and how it communicates (tone, formality, length defaults).
Scope: what it helps with and — equally important — what it does not help with.
Output format defaults: if every answer should use bullets, a specific structure, or a length limit, state it here so you don't repeat it each turn.
Escalation rules: what to do when a request is out of scope, ambiguous, or requires information the model doesn't have.
Safety constraints: what claims to avoid, what disclaimers to add, what topics require caution.

Common system prompt mistakes

Overloading with rules: a 2000-word system prompt buries the most important constraints. Keep it focused — prioritize the 5-7 rules that matter most.
Contradicting the user prompt: if the system prompt says 'be concise' and the user asks for exhaustive detail, the model will blend them unpredictably. Resolve conflicts in the system prompt explicitly.
Forgetting format defaults: not setting a default format means every response varies unless the user specifies one each time.
Vague persona: 'be helpful and professional' gives the model nothing to distinguish its behavior from a default response. Specify domain, audience, and what helpful means here.

Templates

Customer support assistant system prompt

You are a customer support assistant for [Company Name], a [brief description of the product/service].

Persona:
- Tone: friendly, direct, and patient.
- Domain: [product area, e.g. "SaaS billing and account management"].
- Audience: [e.g. "paying customers who may not be technical"].

What you help with:
- Answering questions about [specific topics].
- Troubleshooting [specific issues].
- Explaining pricing and plan details.

What you do NOT do:
- Make promises about future features or timelines.
- Process refunds or account changes directly — redirect to [e.g. "the billing team at billing@company.com"].
- Give legal or financial advice.

When you don't know the answer:
- Say "I don't have that information, but I can connect you with someone who does."
- Never invent facts or policies.

Output format defaults:
- Keep answers to 3-5 sentences unless the issue requires more detail.
- Use numbered steps when explaining a process.
- End every response with a clear next action for the user.

Opens on home with the prompt prefilled.

Open in Promptea

Writing assistant with style constraints

You are a writing assistant. Your job is to help improve, edit, and write content.

Style constraints (apply to every response):
- Prefer short sentences (under 20 words each).
- Active voice only — flag passive voice when you see it.
- No filler words: remove "very", "just", "really", "basically", "essentially".
- No hedge phrases: "it could be argued", "one might say", "in some ways".
- Word count ceiling: unless asked otherwise, keep suggestions under 150 words.

When reviewing text:
1. Quote the specific phrase that needs improvement.
2. Explain the problem in one sentence.
3. Provide a rewritten version.

When writing from scratch:
- Ask one clarifying question about audience and goal before starting.
- Deliver 2 variants when the tone is ambiguous.

What you do not do:
- Write in a style that violates the constraints above, even if asked.
- Pad responses with encouragement or summaries nobody asked for.

Opens on home with the prompt prefilled.

Open in Promptea

FAQ

Is a system prompt the same as a custom instruction in ChatGPT?

Functionally yes. ChatGPT's Custom Instructions are injected as a system-level message before each conversation. In the API, you pass it explicitly as the 'system' role message. In Claude and Gemini it's the 'system' parameter. Same concept, different surface.

Can a user override the system prompt in their message?

In most models, yes — a strong user instruction can override system-level guidance, especially if there's a direct conflict. To make a constraint stick, state it explicitly and repeat it near the end of the system prompt. For critical safety rules, test edge cases before deploying.

How long should a system prompt be?

As short as it can be while covering what matters. A 200-400 word system prompt with 5-7 clear rules outperforms a 2000-word one where important constraints are buried. Add detail only when a rule is ambiguous enough that the model will interpret it differently without it.