Manuscript Reference Auditor

How to use Manuscript Reference Auditor

Find a cited label missing from your reference list, an unused entry, a repeated declaration or a gap in first-mention numbering before you finish a manuscript. Numeric citations and exact labeled footnotes share one transparent source-evidence workflow; the tool does not verify the publications or render a reference style.

1. Paste the manuscript excerpt and put the reference list in its separate field.
2. Choose numeric square-bracket citations or the exact literal footnote profile, then run.
3. Inspect first-mention order, absent labels, unused entries and duplicate declarations without changing your source.

Good to know

• At most 100,000 UTF-16 characters per field, 2,000 reference-list lines, 5,000 recognized groups and 10,000 expanded numeric labels. Numeric labels are canonical integers 1–9,999, preserved as text in the evidence.
• Numeric source supports [1], [1, 3] and ascending hyphen/en-dash ranges of at most 1,000 numbers. Put bibliography entries in label<TAB>nonblank description form. No superscripts, parenthetical styles, locators, citation-manager fields or automatic renumbering is supported.
• Literal footnotes use exact case-sensitive 1–40-character ASCII letter/digit/underscore/hyphen labels. Definitions are separate single lines [^label]: nonblank text; put them in the reference field. This source profile is not a complete GitHub/Markdown footnote renderer.
• Markup, code, escaped-looking brackets and quoted examples are treated literally. Source offsets are zero-based UTF-16; no publication lookup, retraction check, factual citation verification, citation formatting or journal-conformance certification is made.
• Processing is local. Output is limited to 1,000,000 text characters unless a stricter file limit is stated.