Summary
Checks every entry in your BibTeX file against CrossRef, arXiv, and DBLP to catch hallucinated or fabricated citations. You get a three-tier verdict per reference: verified, suspect (field mismatches), or not found. The report breaks down exactly what's wrong at the field level, which is helpful when you're reviewing AI-generated bibliographies or auditing a paper before submission. It's a standalone mode of the OpenJudge PaperReviewPipeline, so you'll need to install py-openjudge and litellm. Adding your email to the CrossRef API calls improves rate limits. The output is a markdown report, and the whole thing runs from the command line without needing to feed it a full PDF.
Install to Claude Code
npx -y skills add agentscope-ai/openjudge --skill bib-verify --agent claude-codeInstalls into .claude/skills of the current project.
Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
An agent that runs the SEO playbooks that move rankings and ships PRs you control.
Get founding access →Files
SKILL.md
Select a file.
Featured
Connect Claude to +800M contacts, +150M companies. Find & Enrich leads in chat.
Try For Free →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →An agent that runs the SEO playbooks that move rankings and ships PRs you control.
Get founding access →First SeenJun 11, 2026
