Back to Glossary
Prompting
System Prompt Best Practices
Definition
System prompt best practices are guidelines for writing effective system-level instructions that define an LLM's behavior, persona, constraints, and output format for consistent, high-quality responses.
Why It Matters
The system prompt is the foundation of your AI application’s behavior. A well-crafted system prompt ensures consistent outputs, appropriate tone, proper handling of edge cases, and safe responses. Poor system prompts lead to unpredictable behavior and user frustration.
Key Best Practices
Structure:
- Start with role/identity definition
- Specify core behaviors and constraints
- Define output format requirements
- Include error handling instructions
- Add safety guardrails
Writing Style:
- Be explicit rather than implicit
- Use positive instructions (“do X”) over prohibitions
- Order instructions by priority
- Include examples for complex formats
Common Patterns
You are [role] that helps users with [task].
Core behaviors:
- Always [key behavior]
- Never [prohibited behavior]
Output format:
- [Format requirements]
If unsure, [fallback behavior].
Test your system prompt with edge cases and adversarial inputs before production deployment.