structured-reference-audit
Skill81 starsChanged 7 days agoMITView on GitHub
Build an auditable reference ledger from a paper PDF with GROBID, then resolve structurally parsed entries through AMiner. Use when a user needs conservative reference-existence checking rather than a direct PDF-level FAKE verdict.
- Reads credentials
- Installs packages
In this repository21 other files it gives its agents.
- aminer-academic-searchSkill
- aminer-daily-paperSkill
- aminer-deep-searchSkill
- aminer-experiment-to-paperSkill
- aminer-exp-extractionSkill
- aminer-free-academicSkill
- aminer-grant-applicationSkill
- aminer-paper-deep-readingSkill
- aminer-paper-writingSkill
- aminer-patent-tech-intelligenceSkill
- aminer-pdf-ocrSkill
- aminer-thesis-writingSkill
- aminer-topic-advisor-selectionSkill
- baseline-finderSkill
- citation-faithfulnessSkill
- deep-researchSkill
- paper-source-traceSkill
- pdf-citation-verifierSkill
- related-work-auditSkill
- scientific-taxonomySkill
- sota-finderSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 15
Use this skill to check whether references can be resolved to scholarly records without confusing PDF parsing failures with fabricated …
- Distinctive: line 62
Do not label an author dishonest from this output.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Structured Reference Audit
Use this skill to check whether references can be resolved to scholarly records
without confusing PDF parsing failures with fabricated citations. It is a
companion to pdf-citation-verifier, not a replacement for AMiner's server-side
PDF verifier.
PDF or GROBID TEI
-> GROBID bibliography entries
-> Reference Ledger: raw entry, page anchor, title/DOI/arXiv, parse warnings
-> AMiner title lookup per eligible entry
-> verified_exists / needs_human_review / not_found_in_aminerRequirements
- Python dependency:
pip install -r requirements.txt - For
--pdf: a running GROBID service, defaulthttp://127.0.0.1:8070. GROBID is external and optional: users with existing TEI may pass--tei. AMINER_API_KEYis required for AMiner resolution, but not for parser-only--skip-resolveruns. Never print or write the token.
Run
python3 scripts/structured_reference_audit.py \
--pdf "/abs/path/paper.pdf" \
--output "reference-ledger.json"For a parser-only diagnostic:
python3 scripts/structured_reference_audit.py \
--tei "/abs/path/paper.tei.xml" \
--skip-resolve \
--output "reference-ledger.json"Interpret Results Conservatively
verified_exists: AMiner returned a high-similarity title match.needs_human_review: parsing, matching, or network evidence is insufficient.not_found_in_aminer: AMiner returned no adequate title candidate. It does not establish that the reference is fabricated.parse_quality_insufficient: do not resolve entries or issue reference-level conclusions until the PDF/TEI parse is inspected.
Do not label an author dishonest from this output. Keep parser fragments in the ledger for inspection and present unresolved entries as a human review queue.
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.

