Contributing

Guidelines

  • Prompts reported in published and peer-reviewed papers only.
  • Prompts from influential preprints may also be considered, provided their effectiveness has been clearly demonstrated.
  • If a prompt has been used in practice (e.g., to assist in specific review tasks), briefly describe the use case and its outcomes.
  • Indicate any limitations or specific requirements for the prompt to work effectively (e.g., tools needed, data formats).
  • Contributions should include information on how the prompt was evaluated (e.g., in academic papers, systematic reviews, or practical use cases).
  • If applicable, include links to resources showcasing the evaluation process or results.

Adding a New Prompt

If you would like to add a new prompt, follow these steps:

  1. Navigate to the page displaying related prompts (e.g., for the search).
  2. Click the Edit this page button.
  3. Add your contribution in Markdown format. Include:
    • The intended use case (e.g., summarization, Boolean query generation).
    • The prompt text.
    • An example output (where feasible).
    • A reference to the academic paper evaluating the prompt to the assets/references.bib.
    • Add a link to the reference (e.g., @SyrianiEtAl2024) brief description of its application and evaluation.
  4. Use the Preview changes tab to ensure your contribution is formatted correctly.
  5. Provide a short name and (optionally) a description for your proposed change.
  6. Select Create a new branch for this commit and start a pull request, then click Propose file change.

Prompt Template

Each prompt should be documented using the following structure. The example below illustrates the expected level of detail and formatting.


### [Short descriptive title of the prompt]

|                        |                                                       |
|------------------------|-------------------------------------------------------|
| **GenAI capability**   | [e.g., Content generation, Data extraction, Analysis] |
| **Prompting strategy** | [e.g., Few-shot prompting, Exploratory prompting]     |
| **Requirements**       | [e.g., LLMs, file upload, large context window]       |
| **Academic study**     | [Citation]                                            |
:  {tbl-colwidths="[30,70]"}

**Prompt**

```Markdown
[Full prompt text as used with the LLM.
Write the prompt exactly as it should be executed,
including role framing, task description, and any
examples or constraints.]
```

**Example output**
*(Include where feasible; omit if not meaningful for the prompt type.)*

> [Representative excerpt of the model output, formatted as a quote.]

Notes for contributors

  • Use clear, descriptive titles (e.g., “Suggest an initial search query”).
  • Prompts should be self-contained and executable.
  • Example outputs should be illustrative, not exhaustive.
  • If no example output is provided, briefly explain why (in the pull request).

License and Contribution Terms

By submitting a contribution to this repository, you agree that:

  • Your contribution is your original work or that you have the right to share it.
  • Your contribution will be made available under the same license as this repository (see LICENSE).
  • You are not transferring copyright; you are granting permission for your contribution to be used, modified, and redistributed under the repository’s license.
  • References to external publications do not imply that those publications are licensed under the repository’s license.

If you have questions about licensing or attribution, please open an issue before submitting a pull request.