Edor.ai Learning Centre
Prompt Engineering Classroom: classroom practice
Eighteen lessons of prompt engineering set entirely in Hong Kong classrooms: planning, tiered worksheets, question setting and rubrics, marking feedback, SEN adaptation, the scaffolded tutor, administrative writing, grounded RAG answers, prompt-injection defence and panel template libraries. Every lesson includes copy-ready prompts in both languages.
18 lessons about 233 minutesFor: classroom teachers, panel heads and teaching assistants; read LLM Classroom modules 1 and 4 first
Start with lesson 1 Syllabus
Module 1: Prompting fundamentals
Frameworks, examples, chain of thought and system prompts
- 01What makes a good promptA real Primary 6 General Studies example showing the gap between a one-line prompt and a structured one, what the four modules of this course cover, and one all-purpose template you can use today. 11 min· Level: Foundation
- 02The role, task, context and format frameworkSplit a prompt into role, task, context and format, add one constraints line, and unpredictable output becomes predictable. Demonstrated on Secondary 3 Mathematics, Secondary 1 English Language and Secondary 5 Visual Arts. 13 min· Level: Foundation
- 03Teaching by example, and output consistencyWhen you have explained the format three times and it still comes back wrong, the problem is not the explanation. Two or three examples lock down item type, feedback register and layout so a whole panel's output looks like one person made it. 12 min· Level: Hands-on
- 04Chain of thought and step-by-step instructionsWrong answers to word problems and long tasks that fall apart halfway are usually not a weak model — they are a model allowed to conclude too early. Three fixes: step-first instructions, a checking pass and task decomposition. 13 min· Level: Intermediate
- 05System prompts and school-based personasA system prompt is a standing instruction that applies to every conversation. Write your school context, language, terminology and safety floor into it once and teachers only ever type the day's task. 13 min· Level: Intermediate
Module 2: Teaching and learning in practice
Planning, question setting, marking, differentiation, the scaffolded tutor and student prompt literacy
- 06Lesson planning promptsPlanning is slow because you start from blank every time, not because you type slowly. Four templates — the plan itself, a slide outline, upgrading an old plan, and post-lesson reflection — to use alongside our 30-minute planning workflow. 14 min· Level: Hands-on
- 07Worksheet and differentiation promptsThe commonest failure in a three-tier worksheet is that the only difference between tiers is the number of questions. Here is how to tier by cognitive level and support instead, with templates for language support and SEN adaptation. 14 min· Level: Hands-on
- 08Question setting and rubric promptsA good paper is derived backwards from the learning focus, not written from whatever comes to mind. Build the table of specifications first, then the items, then the rubric — and finally have AI review the paper you wrote. 15 min· Level: Hands-on
- 09Marking and feedback promptsAI marking belongs in the role of provisional assessment plus specific feedback; no mark it produces may be released without teacher review. Four templates for Chinese composition, English writing, mathematical working and whole-class error patterns, plus the limits on student data. 15 min· Level: Hands-on
- 10Scaffolded tutor prompts that withhold answersA complete tutor system prompt that turns a general AI from an answer machine into a tutor working up four hint levels, with a worked Secondary 3 maths transcript and how the teacher sets the hint ceiling. 13 min· Level: Hands-on
- 11Teaching students prompt literacyA 40-minute lesson you can run as written, with a student worksheet prompt, a method for checking AI claims against original sources, and a specific AI-use declaration students can attach to their work. 12 min· Level: Hands-on
Module 3: Administration, RAG and safety
School documents, grounded answers, prompt injection and student content guardrails
- 12School administration and parent communication promptsWriting administrative prompts with placeholders so identifiable student and staff data never reaches a general-purpose tool, with ready templates for circulars, letters to parents, minutes and a parents' evening script. 11 min· Level: Hands-on
- 13Grounded prompts with citationsConfining the model to the passages you supply, requiring a citation for every claim, and authorising it to answer "not found in the provided documents" — the most effective anti-hallucination technique for school documents. 14 min· Level: Intermediate
- 14Prompt injection and keeping student-facing content safeOne sentence hidden in a composition can move a mark, and hidden text on a web page can change a summary. This lesson gives three defensive patterns — delimiting untrusted content, an instruction hierarchy, and reporting instructions instead of obeying them — and is honest that prompting alone is not enough. 14 min· Level: Intermediate
Module 4: Advanced technique and continuous improvement
Structured output, multimodal prompts, template libraries and a debugging checklist
- 15Structured output as tables and JSONFrom a markdown table to a strict JSON schema — how to specify columns and permitted values, why a fixed structure matters once output feeds a spreadsheet or a system, and what to do when the model wraps commentary around the JSON. 12 min· Level: Intermediate
- 16Multimodal prompts for images, handwriting and audioPrompt patterns for photographing handwritten work, reading a diagram or graph, and giving feedback on English and Putonghua reading aloud — with honest limits on Chinese handwriting recognition and accented speech, and the teacher verification each requires. 12 min· Level: Hands-on
- 17Building a panel prompt template libraryHow a panel names, stores, versions and reviews its prompts so good work stays with the school, including a template metadata format and a termly review cycle. 10 min· Level: Hands-on
- 18A prompt debugging checklistWhen the output is wrong, do not rewrite the whole prompt. Work through six checks in order — is the task ambiguous, is the format unspecified, is the context too long, is the instruction buried, are the examples inconsistent, is the model wrong for the task. 15 min· Level: Intermediate