reference-audit-guide
Skill661 starsChanged 35 days agoMITView on GitHub
Verify that cited references actually exist and that their metadata matches, using the runnable scripts this skill ships for CrossRef, Semantic Scholar, arXiv, and PubMed. Use when the bibliography must be checked against live sources rather than inspected locally: detecting fabricated or hallucinated citations, confirming that DOIs resolve, and matching title, authors, venue, and year against the record. Also covers the underlying verification principles and reference-audit best practices. For an offline hygiene pass over a .bib or .tex file, run citation-verifier first; for whether a real source supports the sentence citing it, run claim-source-verification.
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- Chanw-research/claude-code-paper-writingedited, first seen May 10, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- Chanw-research/claude-code-paper-writing★ 4284% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository26 other files it gives its agents.
- anti-defensive-writingSkill
- citation-verifierSkill
- claim-source-verificationSkill
- data-availabilitySkill
- draft-marker-disciplineSkill
- figure-plannerSkill
- manuscript-optimizerSkill
- paper-bootstrapSkill
- paper-workflowSkill
- rebuttal-responseSkill
- results-section-revisionSkill
- review-article-architectureSkill
- scientific-prose-styleSkill
- scientific-writingSkill
- stats-reporting-auditSkill
- submission-auditSkill
- write-scientific-manuscriptSkill
- figure-styleSkill
- nature-figureSkill
- academic-presentationsSkill
- conference-paper-writingSkill
- academic-researcherSkill
- paper-analyzerSkill
- results-analysisSkill
- paper-reviewerSkill
- nature-portfolio-playbookSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 56
Core idea: Verify immediately when adding a citation, rather than checking after writing is complete.
- Distinctive: line 58
Search for the paper each time a citation is needed
- Distinctive: line 60
Add to bibliography only after verification passes
- Distinctive: line 81
Key principle: When citing a specific claim, confirm the claim actually appears in the paper.
- Distinctive: line 84
Search for relevant keywords
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Citation Verification Reference Guide
Verification principles plus runnable citation checkers for academic writing.
Position in the citation pipeline
This repository separates three citation questions that are routinely confused. Run them in order.
citation-verifier: is the bibliography a well-formed artifact? Duplicate keys, missing fields, DOI syntax, cited-but-undefined. Offline, immediate.- This skill: does the cited work exist, and does its metadata match what the manuscript says? Queries live scholarly APIs. This is the only stage that catches a fabricated citation carrying a well-formed DOI.
claim-source-verification: does the real, correctly-cited source actually support the sentence citing it?
None of the three substitutes for another. A clean local scan shows the bibliography is well-formed, not that the papers are real. A verified DOI shows the paper is real, not that it says what the sentence claims.
Core Principle: Proactively verify every citation during the writing process using trusted scholarly sources rather than memory.
Core Problems
Citation issues in academic papers seriously impact research integrity:
- Fake citations - Citing non-existent papers
- Incorrect information - Mismatched authors, titles, years, etc.
- Inconsistent formatting - Mixed citation formats
- Missing citations - Referenced but uncited work
These issues can lead to:
- Paper rejection or retraction
- Damage to academic reputation
- Reviewers questioning research rigor
Verification Principles
This skill provides verification principles based on trusted scholarly sources:
1. Proactive Verification
Core idea: Verify immediately when adding a citation, rather than checking after writing is complete.
- Search for the paper each time a citation is needed
- Confirm the paper exists in a trusted source
- Add to bibliography only after verification passes
2. Source Hierarchy
Prefer these sources:
- publisher page or DOI resolver
- PubMed for biomedical papers
- arXiv for preprints
- Crossref or Semantic Scholar for metadata cross-checking
- Google Scholar as a fallback discovery aid
3. Information Matching Verification
Information that must match:
- Title
- Authors
- Year
- Publication venue
4. Claim Verification
Key principle: When citing a specific claim, confirm the claim actually appears in the paper.
- Access the paper PDF
- Search for relevant keywords
- Confirm the accuracy of the claim
- Record the section or page where the claim appears
Best Practices
- Never generate citations from memory
- Do not guess when verification fails
- Mark unverifiable references clearly
- Differentiate preprints from published versions
- Verify the claim, not just the metadata
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

