AI Dictionary › Prompting

Skeleton-of-Thought

Skeleton-of-thought asks the model to first produce a skeleton, a terse list of points or headings, then expand each point in parallel or in a second pass. The benefit is twofold: better structure and, when points are expanded in parallel via API, lower latency. Example prompt: "First generate only the bullet outline of the answer (5-7 items, max 8 words each). Then expand each item into a 3-sentence paragraph. Question: how should a new-hire onboarding plan be structured?" The model sketches the structure, then fills the content.

Definition

Use it for long, structured answers, guides, reports, essays, where you want to control structure before detail, or when you can parallelise point expansion to speed generation.

Related terms

• Chain of Thought
• Least-to-Most Prompting
• Output Format
• Prompt Chaining
• Chain-of-Density

More in Prompting

• Prompt
• Prompting
• Role Assignment
• Zero-shot
• Few-shot
• Chain of Thought
• Prompt Engineering
• Output Format

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

📱 Download the Android app (beta) iOS coming soon

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.