Literature review help with real sources and a themed draft
The hardest part of a literature review is the first week: finding the right papers, reading abstracts and seeing how they fit together. This prompt has your todo.is agent search scholarly databases for your topic, give you a table of real papers with links, group them by theme and draft a structured review that you then read, check and rewrite in your own voice.
The prompt
- Help me start a literature review on [RESEARCH QUESTION OR TOPIC] for [PURPOSE AND LEVEL]. Search Google Scholar, PubMed, arXiv, Semantic Scholar and open-access journals for peer-reviewed studies from [YEARS]. Find 15 to 25 relevant papers and only list papers you can link to. Make an Excel table with author, year, title, link, method, sample, main finding and limitations. Then write a Word draft of about [LENGTH] organized by themes, not paper by paper: what is agreed, where findings conflict, and the gaps that need more research. Use [CITATION STYLE] in-text citations and a reference list. Also attach these papers I already have: [MY PAPERS].
What to change
- [RESEARCH QUESTION OR TOPIC]: As specific as you can, e.g. "effects of remote work on employee wellbeing in tech companies".
- [PURPOSE AND LEVEL]: e.g. "a master’s thesis chapter in psychology" or "a grant background section".
- [YEARS]: e.g. "2015 to now" or "the last 10 years".
- [LENGTH]: e.g. "2,000 words".
- [CITATION STYLE]: APA 7, Harvard, MLA, Chicago or Vancouver.
- [MY PAPERS]: Attach PDFs you must include, or write "none".
Example result
- Literature review draft · Remote work and employee wellbeing in tech companies
- Scope: peer-reviewed studies, 2015 to now · 21 papers kept from 64 screened (fictional example)
- How the sources were chosen
- • Searched: Google Scholar, PubMed, Semantic Scholar, open-access journals
- • Kept: empirical studies of knowledge workers; excluded opinion pieces and studies of frontline workers
- • Every paper is in Sources_Table.xlsx with a link, method and sample
- Theme 1: Autonomy and job satisfaction
- Most studies report higher satisfaction when remote work is chosen rather than imposed (Alvarez & Kim, 2019; Osei, 2021). Effects are stronger for employees with long commutes. Samples are mostly from large firms.
- Theme 2: Isolation and loneliness
- Several survey studies link full-time remote work with more loneliness, especially for new hires (Hartmann et al., 2020; Liu, 2022). Hybrid setups show smaller effects. Most designs are cross-sectional, so cause and effect are unclear.
- Theme 3: Work-life boundaries
- Findings conflict. Some studies find better balance (Moreau, 2018), others find longer hours and "always on" behavior (Patel & Novak, 2021). Differences may come from how boundaries were measured.
- Theme 4: Manager support as a moderator
- Three studies suggest supportive managers reduce the negative effects of isolation (Grant, 2020). This is a promising but small body of work.
- Gaps
- • Few longitudinal studies following the same employees over time
- • Little research outside North America and Europe
- • Almost no studies separate fully remote from hybrid work clearly
- Next steps for you
- • Read the 6 papers marked "core" in the table in full
- • Check every claim against the original paper before you cite it
- • Rewrite the draft in your own words and add your own critique
- Note: names and citations above are placeholders for this example. In your real draft, every citation links to a paper your agent actually found.
How to do it with todo.is
- Copy the prompt and add your research question, level, years and citation style.
- Attach any papers you must include, then paste it into todo.is on the Today screen.
- Your agent sends a sources table in Excel and a themed Word draft with references.
- Read the core papers yourself, then ask for changes like "add a methods comparison table" or "focus on studies after 2020".
Tips for a better result
- A narrow question gives a better review. "Remote work and loneliness in new hires" beats "remote work".
- Ask for the screening criteria to be written down. Your supervisor will ask how you chose sources.
- Organize by theme, not by paper. A list of summaries is an annotated bibliography, not a review.
- Verify every citation against the original paper. AI can misread a finding, so never cite a paper you have not opened.
literature review: FAQ
- How many sources should a literature review have? It depends on the level and field. A course paper might use 10 to 20 sources, a master’s chapter 30 to 60, and a PhD review more. Ask your supervisor or check your guidelines.
- Can AI write my literature review for me? It can find sources, summarize them and draft a structure, which saves a lot of time. You still need to read the key papers, check every claim and write the final version yourself, following your school’s rules on AI use.
- Will the agent invent fake references? The prompt tells it to list only papers it can link to, and every source in the table has a link. Still open each one to confirm the details before you cite it.
- Can it read paywalled journal articles? Only open pages. If you have access through your library, download the PDFs and attach them so your agent can read the full text.
JavaScript is required to use the todo.is app.