Prompt Engineering Fundamentals / Prompt Optimization

Clarity

Essential [1/5]
Explicitness Unambiguity

Definition

Clarity in prompt engineering means making instructions unambiguous and specific so the LLM clearly understands what is expected. A clear prompt leaves no room for misinterpretation.

Ambiguous prompts force the model to guess intent, often leading to outputs that don't match expectations. Clear prompts dramatically improve output quality and consistency.

Key Concepts

  • Explicit intent: State exactly what you want, not what you don't want
  • Defined terms: Avoid jargon or explain specialized vocabulary
  • Single interpretation: Each instruction should have only one meaning
  • Concrete examples: Show what success looks like

Examples

Unclear vs Clear
Task Specification
Unclear: "Fix this code." Clear: "Fix the syntax error in line 5 of this Python function. The function should return the sum of two numbers. Keep the function name and parameter names unchanged."
The clear version specifies what to fix, what the expected behavior is, and what to preserve.
Ambiguous vs Unambiguous
Output Expectations
Ambiguous: "Make it shorter." Unambiguous: "Reduce this paragraph to 50 words or fewer while retaining the main argument about climate change and the call-to-action at the end."
"Shorter" could mean many things; specifying word count and required elements removes ambiguity.

Interactive Exercise

✎
Improve Clarity

Rewrite this unclear prompt to be more clear:

Unclear: "Write something about dogs."

Pro Tips
  • Read your prompt aloud—if it sounds vague, it probably is
  • Ask "could this be interpreted differently?" for each sentence
  • Use numbers instead of vague quantifiers ("3 examples" not "a few examples")
  • When in doubt, over-specify rather than under-specify

Related Terms

Specificity Instruction Following Prompt Engineering Constraint Specification