bib-verify
Skill3 starsChanged 8 months agoMITView on GitHub
Verify BibTeX entries against CrossRef API. Checks DOIs, titles, authors, journals, years for accuracy. Detects fabricated or incorrect entries. Use when auditing .bib files for a thesis or paper. Chinese triggers: 参考文献, bib, 引用检查, DOI, 论文引用, 文献校验.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- acprk/crypto-research-agentstheir own version, first seen Nov 4, 2026
- rizavelioglu/bib-verifytheir own version, first seen Aug 2, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- acprk/crypto-research-agents★ 15same name
- rizavelioglu/bib-verify★ 0same name
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 repository8 other files it gives its agents.
- ace-searchSkill
- clashverge-rules-debugSkill
- grok-imagineSkill
- grok-searchSkill
- hydraSkill
- lit-reviewSkill
- memory-palaceSkill
- readmeSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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 8
Verify BibTeX references for: $ARGUMENTS
- Distinctive: line 10
If no file path is given, ask the user for the .bib file path.
- Distinctive: line 16
Read the .bib file and extract all entries.
- Distinctive: line 57
If a match is found, report the real DOI so it can be added.
- Distinctive: line 58
If no match is found, flag the entry as unverifiable (may need manual check).
- Distinctive: line 86
OK: All verified fields match
≈ “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
Instructions
Verify BibTeX references for: $ARGUMENTS
If no file path is given, ask the user for the .bib file path.
Step 1: Extract entries from the bib file
Read the .bib file and extract all entries. For each entry, collect:
- Citation key
- DOI (if present)
- Title
- Authors
- Journal/Booktitle
- Year
- Volume, pages
Step 2: Verify entries with DOI
For entries that have a DOI, verify via CrossRef API:
curl -s "https://api.crossref.org/works/{DOI}" --max-time 8 | python3 -c "
import sys,json
d=json.load(sys.stdin)['message']
print('Title:', d['title'][0][:100])
print('Authors:', ', '.join([a.get('family','') for a in d.get('author',[])][:5]))
print('Year:', d.get('published',{}).get('date-parts',[['']])[0][0])
print('Journal:', d.get('container-title',[''])[0][:60])
print('Volume:', d.get('volume',''))
print('Pages:', d.get('page',''))
"Compare the returned metadata with the bib entry. Flag mismatches in:
- Title (significant difference, not just capitalization)
- Authors (different people, not just name format)
- Journal name (completely different journal)
- Year (off by more than 1 year)
- Volume or pages (significant difference)
Step 3: Verify entries without DOI
For entries without DOI, search CrossRef by title:
curl -s "https://api.crossref.org/works?query={URL_ENCODED_TITLE}&rows=1" --max-time 8If a match is found, report the real DOI so it can be added. If no match is found, flag the entry as unverifiable (may need manual check).
Step 4: Batch processing
Process DOIs in batches of 5-8 to avoid rate limiting. Use parallel curl calls where possible.
For large bib files (>50 entries), first extract all DOIs:
grep -oP 'doi\s*=\s*\{([^}]+)\}' refs.bib | sed 's/doi\s*=\s*{//;s/}//'Step 5: Report
Output a summary table:
| Key | Status | Issue |
| key1 | OK | - |
| key2 | MISMATCH | Title differs: bib says "X", CrossRef says "Y" |
| key3 | DOI_NOT_FOUND | DOI does not exist in CrossRef |
| key4 | NO_DOI | No DOI provided, searched by title, found DOI: 10.xxx |
| key5 | FABRICATED | Journal/authors completely wrong |
Severity levels:
- FABRICATED: Journal, authors, or title entirely wrong (GPT hallucination)
- MISMATCH: Some fields incorrect but paper exists
- DOI_NOT_FOUND: DOI returns 404/no result
- NO_DOI: Entry has no DOI (suggest adding one)
- OK: All verified fields match
Notes
- CrossRef API is free and doesn't require authentication
- Rate limit: ~50 requests/second for polite pool (add
mailtoparam for higher limits) - IEEE DOIs often return HTTP 418 (anti-bot) on doi.org but work fine on CrossRef API
- Books (@book) may not be in CrossRef; flag as "unverifiable" rather than "fabricated"
- Conference papers (@inproceedings) may have limited metadata in CrossRef
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.

