How to Write Better Prompts: The TRACE Framework
Discover the TRACE method (Task, Role, Audience, Context, Examples) for writing AI prompts that give you high-quality, classroom-ready results every time.
Video coming soon
Add a YouTube URL in Airtable or data/tutorials.ts
About this tutorial
Most AI prompts fail not because the AI is bad, but because the instruction is incomplete. TRACE is a five-part framework that ensures your prompts give the AI everything it needs to produce something genuinely useful.
T — Task: what exactly do you want the AI to produce? R — Role: what role should the AI adopt (e.g. "experienced IELTS examiner")? A — Audience: who is this for (e.g. "B1 adult learners in a Business English context")? C — Context: what's the situation or topic? E — Examples: show the AI what good looks like.
In this tutorial we walk through three before/after examples — a grammar exercise, a speaking activity, and a writing feedback prompt — showing how applying TRACE transforms each one from mediocre to classroom-ready.
Prompts used in this tutorial
Grammar Error Correction Worksheet
You are an EFL grammar teacher. Create a grammar error correction exercise for [CEFR LEVEL] students focusing on [GRAMMAR POINT, e.g. present perfect vs. past simple]. Write 10 sentences, each containing exactly one grammatical error related to the target grammar point. Use natural, contextually rich sentences — avoid isolated, artificial examples. After the exercise, provide an answer key with a brief explanation for each correction. Topic theme for the sentences: [TOPIC].