Back to Lessons
Integrated Course

Prompting Best Practices

Claude 4.5 field guide for explicit instructions, context-aware workflows, and agent orchestration

Chapter 11: General principles (Claude 4.5)

Learning Content

Claude 4.5 models reward explicit, context-rich instructions. State desired outputs, stretch goals, and guardrails up front. When you explain why constraints matter, Sonnet and Haiku follow them more reliably.

Key Techniques

Be explicit about outputs, constraints, and stretch goals

Add motivation/context so rules feel purposeful

Align examples with the behavior you want reproduced

Spell out role, audience, format, and evaluation criteria

Common Pitfalls

Assuming Claude will infer missing context or audience

Providing examples that conflict with desired behavior

Leaving quality bars and guardrails implicit

Prompt Examples

Less effective: Create an analytics dashboard | More effective: Create an analytics dashboard. Include as many relevant features and interactions as possible. Go beyond the basics to create a fully-featured implementation.
Less effective: NEVER use ellipses | More effective: Your response will be read aloud by a text-to-speech engine, so never use ellipses since the engine will not know how to pronounce them.
Clarify tone and scope: Explain the audience, success criteria, and any anti-goals so Claude can optimize for them.
1 / 5
Prompting Best Practices - Claude 4.5 Field Guide | 12Factor.me