Academic writing is a long pipeline — search strategy, literature synthesis, abstract, methods, discussion, and the bruising peer-review cycle. This pack gives you tested prompt templates for each stage, each with fill-in-the-blank variables and citation guardrails so the AI helps you structure and clarify without inventing the scholarship.
How it works
Each prompt names your field, paper type, and the section you’re working on, and asks
the model for structure, framing, and language — never for sources or facts to be trusted. The
bracketed variables — [field], [paper type], [section needed] — shape the output to your
discipline. Filter by category or search, replace the brackets, copy, and paste into your AI tool.
Why citation guardrails matter
Language models routinely fabricate references — real-looking author names, titles, journals, and DOIs that do not exist. Every prompt here is written so the model works only with notes and sources you provide, or helps you design searches you then run yourself in a real database. Never let an AI supply a citation you have not personally located and verified.
Tips for better output
- Supply your own sources. Paste your notes and ask the model to synthesise — do not ask it to “find papers”, which produces hallucinated references.
- Use the self-review prompt before submitting. Asking the model to attack your work as a tough reviewer surfaces weaknesses you can fix cheaply in advance.
- Keep your voice and your analysis. Use AI for structure and clarity; the interpretation, argument, and originality must be genuinely yours.
- Follow your institution’s AI policy. Disclose use where required, never list AI as an author, and keep unpublished data and confidential manuscripts out of public AI tools.