A specialized tool for auditing .bib files to detect hallucinated or incorrect academic references by cross-referencing global databases.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install bib-verify
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install bib-verify using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is BibTeX Verification Skill?
The BibTeX Verification Skill is a robust auditing tool designed to combat the growing issue of AI-generated hallucinations in academic writing. By leveraging the OpenJudge PaperReviewPipeline, this skill meticulously parses .bib files and validates every entry against authoritative sources including CrossRef, arXiv, and DBLP.
This utility is essential for researchers and developers using Openclaw Skills who need to ensure that their bibliographies are accurate, verifiable, and free of fabricated data. It provides granular field-level analysis to distinguish between minor formatting errors and completely non-existent citations.
BibTeX Verification Skill Use Cases
- Validating bibliography files generated by LLMs to catch potential hallucinations.
- Auditing academic papers before submission to ensure all DOIs and metadata are correct.
- Checking historical .bib databases for broken links or outdated publication years.
- Automating the quality assurance process for large-scale literature reviews.
How BibTeX Verification Skill Works
- Parsing: The skill extracts all metadata entries from the provided .bib file.
- Database Querying: It performs simultaneous lookups across CrossRef, arXiv, and DBLP databases.
- Comparison Engine: The tool compares titles, authors, years, and DOIs between the local file and the official database records.
- Status Assignment: Each entry is categorized as verified, suspect, or not_found based on matching confidence.
- Reporting: A detailed Markdown report is generated, highlighting specific field-level mismatches.
BibTeX Verification Skill Setup
To get started with this skill, install the necessary dependencies via pip:
pip install py-openjudge litellm
You can then run the verification directly from your CLI:
python -m cookbooks.paper_review --bib_only references.bib --email [email protected] --output report.md
BibTeX Verification Skill Data Schema & Taxonomy
The skill produces a structured report based on the following classification taxonomy:
| Status | Definition |
|---|---|
| verified | Match found in database with consistent field data. |
| suspect | Entry exists but significant field mismatches (e.g., title/author) suggest a mis-citation. |
| not_found | No database match found; entry is likely fabricated or hallucinated. |
Field Mismatch Tracking:
- title_match (Boolean)
- author_match (Boolean)
- year_match (Boolean)
- doi_match (Boolean)
BibTeX Verification Skill Advanced Features
- Rate Limit Optimization: Supports CrossRef mailto integration to prioritize API requests.
- Multi-Language Reporting: Generate verification audits in both English (en) and Chinese (zh).
- Field-Level Forensics: Detailed breakdown of why a citation failed, helping users identify if an error is a typo or a hallucination.
- Openclaw Skills Integration: Can be combined with the full PDF review pipeline for end-to-end manuscript validation.
SKILL.md
Loading
Related Openclaw Skills
CPA Calculator Ecommerce
A powerful financial modeling tool to determine acquisition cost guardrails for ecommerce brands before scaling ad spend.
Dependency Upgrade Briefing
A specialized AI agent skill designed to analyze and explain the benefits, risks, and rollback strategies of software dependency updates.
Tukeli Visual API
A comprehensive AI image processing toolset for automated background removal, facial clarity enhancement, and AI-generated background replacement.
Baidu Qianfan Search
A professional-grade integration for the Baidu Qianfan Enterprise Search API, enabling AI agents to perform advanced web, image, and video searches in Chinese.
Screenshot OCR
A high-precision OCR utility that transforms screenshots and clipboard images into editable text, supporting multiple languages and formats.
Mugua Legal Consultation
A professional AI legal assistant that provides expert consultation, automated case element extraction, and comprehensive legal analysis via the Mugua API.
Featured*
