Prompt Workshop
Why it is written this way
Tap a block to see its paragraph
You are a experienced teaching assistant for {{academic subject}} who has graded student assignments across multiple semesters. You know exactly what students tend to overlook and where they lose points.
Below is the full text of the assignment prompt. The required length is {{content amount}}, and I am not sure where to start.
Before generating the outline, please ask any clarifying questions if key details cannot be determined from the prompt alone. Limit your questions to a maximum of three high-impact items that would significantly alter the structure, and wait for my response before creating the outline.
Provide the final outline in four parts: 1) A requirements checklist extracted from the notice (including submission format, deadline, and length requirements where missing points will lead to deductions). 2) Expected grading criteria along with the exact phrases from the prompt that justify them. 3) A section-by-section outline with suggested length/word count allocation for each part. 4) A step-by-step action plan with estimated time requirements for each phase.
Do not invent requirements not mentioned in the prompt. If you make an assumption about an unstated detail, explicitly flag it as "Not specified in prompt – Confirm with professor." Do not write the actual essay; only generate the outline and plan.
Assignment notice: """ {{assignment notice}} """
- Role Who answers ×1
- Material What the AI needs to know ×2
- Work What to do, and how ×1
- Output What comes back, in what shape ×2
The most common mistake when starting an assignment is skimming the prompt and diving straight into writing. Students often realize requirements like "at least 3 academic sources" or "attach plagiarism report" at the last minute, leading to rushed fixes and lost points. This prompt converts raw assignment text into an actionable checklist before writing begins.
Read the full explanation · 3 paragraphs
Placing clarifying questions before the outline is the core technique here. Assignment prompts are often brief and leave out details like the scope of acceptable companies or citation formats. Generating an outline without clarifying these leads to generic, mismatched structures. Limiting it to three questions prevents the AI from stalling the workflow with an endless questionnaire.
Specifying the context of the academic subject and the TA role sets the standard for expected output quality—an "analysis" in a business course demands different evidence than one in history. In the four-part format, requiring AI to quote specific phrases for each grading criterion prevents it from hallucinating generic rubrics.
Finally, the concluding constraints prevent hallucinated requirements and stop the AI from drafting the paper itself. Wrapping the assignment prompt in """ ensures the AI treats instructional phrases like "Analyze..." as source data rather than direct instructions to execute immediately.
Unfamiliar terms? See Aha AI: output-format, prompt
Compared with a bad example
Make an outline for my assignment. It's for Principles of Management and I need to apply a lecture concept to a real company.
Summarizing the prompt yourself strips away crucial constraints. Length, deadline, and source requirements—the easiest places to lose points—get omitted, and without knowing specific lecture topics, the AI will produce a generic outline. You end up having to re-read the original prompt and start over anyway.
Variations
When dividing tasks for a group project
Group projects often stall at handoff points. Adding an explicit "Overlap/Handoff Points" column surfaces dependencies early.
Show prompt
Please review the {{academic subject}} assignment prompt below and break it down for a 4-person team. First, create a requirements checklist. Then, provide a table with columns for "Assigned Section | Key Tasks | Estimated Time | Overlap/Handoff Points with Others." Include a final integration step and specify how many days before the deadline the team should assemble the final draft.
""" {{assignment notice}} """
When doing a final pre-submission check
Right before submission, you need a checklist rather than an outline. This variant prioritizes high-penalty compliance items first.
Show prompt
Extract all compliance conditions from the assignment prompt below and convert them into a final pre-submission checklist. Write each item as a clear "Yes/No" question, prioritizing high-risk deduction items at the top (formatting, length requirement: {{content amount}}, deadline, required attachments). Group any ambiguous requirements in a separate section.
""" {{assignment notice}} """
Model notes
Some models may jump straight to generating an outline despite being instructed to ask questions first. If this happens, follow up with: "Please hold off on the outline and ask your top clarifying questions first." If your assignment prompt contains tables or images, convert the text and paste the table contents cleanly.
Last updated 2026-09-02 · Found a mistake? Let us know