WoluTools
Find a tool

Education & Research

Check that every in-text citation matches your reference list

Upload a DOCX, PDF, TXT or BIB file. You get citations with no reference entry, references never cited, and mismatched or inconsistent entries, each with its location.

or drop it here

DOCX, PDF, TXT, BIB · Up to 4,000 words · 64,000 extracted characters maximum
AI tool: needs an account and credits, cost shown before you start

No file at hand? See the prepared example

AI job1 credit = 1 small AI job1 AI creditper job · Pro includes 40 credits a month · packs from €6.99
  • DOCX / PDF / TXT / BIB
  • Up to 4,000 words
  • 64,000 extracted characters maximum
Prepared result preview · fictional sample dataFictional paper excerpt
Source
Paragraph 4: Prior work supports this model (Meyer, 2024).
References
Stone, A. (2022). Sample Systems.
Prioritized findings
  • Citation without listed reference: Meyer, 2024
  • Orphan candidate: Stone, A. (2022)

Citation review workbook, mismatch CSVs, format findings and evidence HTML

BringDOCX, PDF, TXT, BIB
GetCitation review workbook, mismatch CSVs, format findings and evidence HTML
PrivacyEncrypted source · 24-hour result

One clear job, from source to download

  1. 1

    Add the source

    Supported formats and limits are visible before the upload.

  2. 2

    Confirm the settings

    Review the exact source, options, units and access before processing.

  3. 3

    Inspect and download

    Check the preview and warnings, then unlock the complete package.

Match in-text citations against your reference list

What you upload

You upload a DOCX, PDF, TXT or BIB file of up to 4,000 words and 64,000 extracted characters. A scanned PDF without a text layer will not extract. For a long thesis, check one chapter at a time. This is an AI tool. It needs an account and costs 1 credit per job, and the cost is shown before you start.

What the results show

You get citations that have no entry in the reference list, and listed entries that are never cited. Formatting inconsistencies are listed too. Every finding keeps the literal citation text and its location, so you can find the place in your document. Two works by the same author and year are kept apart and flagged, not merged. The results come as a workbook, CSV files and an HTML evidence page.

Citation style and your final check

You can set the style to APA 7, Harvard, Chicago author-date, IEEE or Vancouver, or let the tool detect it. The setting shapes how formatting is judged, not whether a citation is found. The findings are candidates from extracted text, not a pass or fail verdict. You decide which ones are real errors and correct them yourself.

Questions before you run it

What counts as a missing or an orphaned reference?

A missing reference is an in-text citation with no matching entry in the bibliography. An orphan is a listed entry that is never cited in the body. Both come back with the literal string and its location.

Which citation styles does it understand?

APA 7, Harvard, Chicago author-date, IEEE and Vancouver, or auto-detect when you are unsure. The style setting shapes how formatting inconsistencies are judged, not whether a citation is found.

Is this the same as a CSL or BibTeX validator?

No. It returns candidate mismatches from extracted text rather than validating against a style specification, so the findings are a review list you work through, not a pass or fail verdict.

What happens with two papers by the same author and year?

Ambiguous author-year matches are kept apart and flagged instead of merged, so you decide whether Meyer 2024a and Meyer 2024b were cited correctly.

How long a document can I check in one run?

Up to 4,000 words, with a ceiling of 64,000 extracted characters. A scanned PDF with no text layer will not extract, so a chapter at a time works better than a whole thesis.