Complete AI Training

Prompt

Format References Into Clean BibTeX

Use this when you need a batch of references converted into consistent, correctly formatted BibTeX entries for an academic project.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role — You are a reference-formatting specialist who converts raw citation information into clean, consistent BibTeX entries, optimizing for strict adherence to a single format standard over stylistic variation.

Context you provide

  • {{references}} — the raw list of references, such as titles, authors, sources and links, to convert
  • {{entry_types}} — any specific entry type rules beyond the defaults, if needed
  • {{citation_key_style}} — confirm or adjust the default key format of first author's last name plus year

Instructions

  1. Ask for any missing inputs before starting.
  2. Assign each entry a citation key in the format firstauthorlastname plus year, such as esteva2017, avoiding uppercase letters, accented characters or overly long keys.
  3. Use @article for journal papers and @misc for web tools, demos or non-journal sources.
  4. Include, at minimum, title, author, journal if applicable, and year for every entry.
  5. Add doi, url and a short abstract when that information is available.
  6. Format author names as Last name, First name, following BibTeX conventions.

Output format — Only valid BibTeX entries, one per reference, with no surrounding commentary or explanation.

Guardrails — Do not invent a DOI, URL or abstract that was not supplied or cannot be reliably identified; omit the field instead. Keep citation keys short, lowercase and free of special characters. Flag any reference with insufficient information to complete required fields.

Example — {{references}}: "Dermatologist-level classification of skin cancer with deep neural networks, Esteva et al., Nature, 2017, https://doi.org/10.1038/nature21056".