bib-verify

Verify a BibTeX file for hallucinated or fabricated references by cross-checking every entry against CrossRef, arXiv, and DBLP. Reports each reference as ver...

Security Scan

The skill's stated purpose matches its instructions, but it asks you to install and run third‑party Python packages (not provided or pinned) without declaring required binaries or credentials — that gap is risky and unexplained.

Purpose & Capability

The SKILL.md describes verifying .bib entries against CrossRef, arXiv, and DBLP which is coherent with the skill name and description. However the manifest declares no required binaries while the runtime instructions assume a Python runtime and pip (e.g., `python -m ...`, `pip install ...`), an inconsistency the publisher did not declare.

Instruction Scope

The runtime instructions direct the agent/user to install and execute a Python module (`cookbooks.paper_review`) that is not included in the skill bundle. Running that module will execute arbitrary third‑party code and network queries; the SKILL.md does not show the implementation or indicate what the module does beyond high‑level behavior.

Install Mechanism

There is no formal install spec in the registry. The README suggests `pip install py-openjudge litellm`, which instructs installing packages from PyPI (or another index) at runtime. Those packages are not pinned to versions nor provided, so you would be pulling and running external code without an audit trail.

Credentials

The skill declares no required environment variables or credentials (CrossRef email is optional). However one dependency is `litellm` which may require LLM API credentials or network access depending on configuration; the SKILL.md does not declare these requirements or how secrets (if any) are used.

Persistence & Privilege

The skill does not request permanent presence (always:false) and does not declare modifications to other skills or system settings.

What to consider before installing

This is an instruction-only skill that tells you to install and run external Python packages not included in the bundle. Before installing or running: (1) verify the exact PyPI package names and inspect their source code/release pages; (2) prefer installing in an isolated environment (virtualenv or container) and pin specific package versions; (3) be aware the tool will make network requests to CrossRef/arXiv/DBLP and may require LLM-related credentials depending on `litellm` usage — confirm what credentials (if any) are needed and why; (4) if you cannot review the packages' code, treat the operation as higher risk and consider asking the publisher for the implementation or a signed release.

BibTeX Verification Skill

Check every entry in a .bib file against real academic databases using the OpenJudge PaperReviewPipeline in BibTeX-only mode:

1. Parse — extract all entries from the .bib file
2. Lookup — query CrossRef, arXiv, and DBLP for each reference
3. Match — compare title, authors, year, and DOI
4. Report — flag each entry as verified, suspect, or not_found

Prerequisites

pip install py-openjudge litellm

Gather from user before running

Info	Required?	Notes
BibTeX file path	Yes	.bib file to verify
CrossRef email	No	Improves CrossRef API rate limits

Quick start

# Verify a standalone .bib file
python -m cookbooks.paper_review --bib_only references.bib

# With CrossRef email for better rate limits
python -m cookbooks.paper_review --bib_only references.bib --email [e-mail]

# Save report to a custom path
python -m cookbooks.paper_review --bib_only references.bib \
  --email [e-mail] --output bib_report.md

Relevant options

Flag	Default	Description
--bib_only	—	Path to .bib file (required for standalone verification)
--email	—	CrossRef mailto — improves rate limits, recommended
--output	auto	Output .md report path
--language	en	Report language: en or zh

Interpreting results

Each reference entry is assigned one of three statuses:

Status	Meaning
verified	Found in CrossRef / arXiv / DBLP with matching fields
suspect	Title or authors do not match any real paper — likely hallucinated or mis-cited
not_found	No match in any database — treat as fabricated

Field-level details are shown for suspect entries:

• title_match — whether the title matches a real paper
• author_match — whether the author list matches
• year_match — whether the publication year is correct
• doi_match — whether the DOI resolves to the right paper

Additional resources

• Full pipeline options: ../paper-review/reference.md
• Combined PDF review + BibTeX verification: ../paper-review/SKILL.md