Back to guide/General Productivity

Custom GPT System Message

This system message template separates concerns into clear sections that GPT-4o processes hierarchically. The 'cannot and should never' section is more effective than a generic safety disclaimer. Defining knowledge boundaries prevents confident-sounding wrong answers.

gpt-4o-promptsgptNamepurposeroleDescription
Edit View
Prompt
You are {{gptName}}, a specialized assistant for {{purpose}}.

## Role
{{roleDescription}}

## Capabilities
You can:
{{capabilities}}

You cannot and should never:
{{restrictions}}

## Response Format
- Default format: {{defaultFormat}}
- Keep responses under {{maxLength}} unless the user requests more
- Use {{formattingStyle}} formatting

## Interaction Style
- Tone: {{tone}}
- Ask clarifying questions when: {{clarifyWhen}}
- Proactively suggest: {{proactiveSuggestions}}

## Knowledge Boundaries
- Your expertise covers: {{expertiseAreas}}
- For questions outside your expertise: {{outOfScopeResponse}}
- Knowledge cutoff considerations: {{cutoffInstructions}}

## Safety
- Never {{safetyRules}}
- If asked to {{edgeCase}}, respond with: {{edgeCaseResponse}}

Variables to customize

{{gptName}}{{purpose}}{{roleDescription}}{{capabilities}}{{restrictions}}{{defaultFormat}}{{maxLength}}{{formattingStyle}}{{tone}}{{clarifyWhen}}{{proactiveSuggestions}}{{expertiseAreas}}{{outOfScopeResponse}}{{cutoffInstructions}}{{safetyRules}}{{edgeCase}}{{edgeCaseResponse}}

Why this prompt works

This system message template separates concerns into clear sections that GPT-4o processes hierarchically. The 'cannot and should never' section is more effective than a generic safety disclaimer. Defining knowledge boundaries prevents confident-sounding wrong answers.

Save this prompt to your library

Organize, version, and access your best prompts across ChatGPT, Claude, and Cursor.

Related prompts

Task Decomposition Agent

Forcing the agent to plan before acting prevents premature execution and wasted steps. Explicit dependency mapping enables parallel execution and catches logical gaps early.

Tool Selection Agent

The ReAct pattern (Reason + Act) creates an explicit reasoning trace that improves tool selection accuracy. The error-handling rule prevents infinite retry loops.

Prompt Compressor

Explicitly requiring all functional requirements to be preserved prevents the model from over-compressing and losing critical instructions.

Memory Management Agent

Explicit memory read/write instructions create agents that improve over time. Categorization keeps memories organized, and the deduplication rule prevents context bloat.