bibguard 0.3.1

Detect hallucinated and broken citations in academic papers

bibguard

Detect hallucinated and broken citations in academic papers.

One command to verify every reference in your .bib file against five scholarly databases. Catches phantom DOIs, fabricated arXiv IDs, author mismatches, retracted papers, and AI-hallucinated citations.

pip install bibguard        # Python
npx bibguard paper.bib      # Node.js (zero install)

Landing Page | PyPI | npm | Browser Extension | Changelog

---

Why

Large language models hallucinate citations. Copy-paste errors corrupt metadata. Retracted papers slip through review. bibguard catches these problems before submission.

• 5 sources: arXiv, Crossref, DBLP, Semantic Scholar, OpenAlex
• Phantom ID detection: Valid-format DOI/arXiv that doesn't resolve = hallucination signal
• Kill-shot logic: A phantom ID cannot be overridden by a similar search result
• TeX cross-audit: Find \cite{key} with no .bib entry, and orphan entries never cited
• Duplicate detection: Flag near-identical entries with different keys
• Auto-fix: Generate a corrected .bib with missing DOIs and eprint IDs filled in
• Type-aware: @misc/@online entries won't false-alarm as "hallucinated"
• Zero heavy dependencies: Core requires only requests + bibtexparser

Install

# Python
pip install bibguard            # minimal
pip install bibguard[fast]      # + RapidFuzz for better title matching
pip install bibguard[all]       # + RapidFuzz + PyMuPDF for PDF parsing

# Node.js / TypeScript (zero dependencies)
npx bibguard paper.bib          # run directly
npm install bibguard             # as library

# Browser extension
# Download from https://github.com/GeoffreyWang1117/bibguard-ext

Python requires 3.9+. Node.js requires 18+.

Usage

CLI

# Basic: verify all entries in a .bib file
bibguard references.bib

# With TeX cross-audit (finds phantom \cite and orphan entries)
bibguard references.bib --tex main.tex

# Save report + auto-fix
bibguard references.bib --tex main.tex --out report.md --fix fixed.bib

# Parallel verification (auto-selects workers for large files)
bibguard references.bib -w 4

# JSON output (for CI pipelines)
bibguard references.bib --json --out report.json

Python API

from bibguard import verify_bib, verify_entry

# Verify entire .bib file (parallel with 4 workers)
results, report = verify_bib("references.bib", tex_path="main.tex", workers=4)
for r in results:
    if r.overall != "OK":
        print(f"{r.overall}: {r.key} -- {r.title}")

# Verify a single entry
from bibguard.parsers.bibtex import parse_bib
entries = parse_bib("references.bib")
result = verify_entry(entries[0])
print(result.overall, result.checks)

TypeScript / Browser

import { parseBib, verifyAll } from "bibguard";

const entries = parseBib(bibText);
const results = await verifyAll(entries, (i, total, key, status) => {
  console.log(`[${i}/${total}] ${key}: ${status}`);
});

All 5 APIs support CORS — works directly in the browser without a proxy.

Exit codes

Code	Meaning
0	All entries OK or WARN
1	At least one FAIL
2	Input error (file not found)

Benchmark

Golden test set (58 cases)

Reproduce with python tests/bench_golden.py.

Category	Metric	Result
Hallucinated (14 fabricated)	Detected as FAIL	14/14 (100%)
Chimera (5 mixed-metadata)	Detected as >= WARN	5/5 (100%)
Real papers (10 legitimate)	False positive (FAIL)	0/10 (0%)

Large-scale validation (200 cases)

Sampled from crawled datasets (800 hallucinated, 400 chimera, 656 real, 153 retracted). Reproduce with python tests/bench_large.py.

Category	N	OK	WARN	FAIL	Key metric
Hallucinated	50	0	0	50	100% detected (all FAIL)
Chimera	50	0	18	32	100% detected
Real papers	50	43	7	0	86% clean, 0% false positive
Retracted	50	49	1	0	2% flagged (L0 limitation)

For semantic NLI, citation graph analysis, and Bayesian risk scoring, see IntegriRef.

AI Coding Assistant Integration

bibguard ships with skill/rule definitions for major AI coding assistants.

Claude Code

mkdir -p ~/.claude/commands
curl -o ~/.claude/commands/bibguard.md \
  https://raw.githubusercontent.com/GeoffreyWang1117/bibguard/main/.claude/commands/bibguard.md

Then use /bibguard paper.bib in Claude Code.

OpenAI Codex CLI

mkdir -p ~/.codex/skills/bibguard
curl -o ~/.codex/skills/bibguard/SKILL.md \
  https://raw.githubusercontent.com/GeoffreyWang1117/bibguard/main/.codex/skills/bibguard/SKILL.md

Cursor

mkdir -p .cursor/rules
curl -o .cursor/rules/bibguard.md \
  https://raw.githubusercontent.com/GeoffreyWang1117/bibguard/main/.cursor/rules/bibguard.md

Any other assistant

bibguard paper.bib --json --out report.json

API sources

Source	Lookup method	CORS	Coverage
arXiv	ID resolution	Yes	CS, Physics, Math
Crossref	DOI resolution	Yes	150M+ records
DBLP	Title search	Yes	CS papers
Semantic Scholar	Title search	Yes	200M+ papers
OpenAlex	Title search	Yes	250M+ works

All queries respect rate limits. No API keys required.

Contributing

git clone https://github.com/GeoffreyWang1117/bibguard.git
cd bibguard
pip install -e ".[dev]"
pytest

Related

• IntegriRef -- Full L0-L4 verification stack with semantic NLI (93.5%), citation graph analysis, and Bayesian risk scoring
• bibguard-js -- TypeScript version (zero deps, browser-native)
• bibguard-ext -- Chrome/Firefox browser extension

Contributors

See CONTRIBUTORS.md for detailed attribution.

• Geoffrey Wang -- Architecture, core algorithms, phantom-ID detection, kill-shot logic, benchmark design
• Claude (Anthropic) -- Modular refactoring, output formatting, packaging, documentation

Support

If you find bibguard useful, please consider giving it a star on GitHub — it helps others discover the project.

Have a bug report, feature request, or suggestion? Open an issue — all feedback is welcome.

Roadmap

• Parallel verification — Shipped in v0.3.1: --workers N flag for concurrent verification via thread pool
• Async I/O — migrate from requests to asyncio + aiohttp for further speedup on 500+ entry files
• Batch API queries — leverage Crossref, Semantic Scholar, and OpenAlex batch endpoints to reduce per-entry overhead
• Caching layer — local cache for repeated lookups across runs
• Retracted paper detection — integrate Retraction Watch database for L0-level retraction flagging

License

Apache License 2.0. See LICENSE for details.

---

中文说明

bibguard — 学术论文引用幻觉检测工具

一行命令，检测论文中的虚假引用。 Python + TypeScript + 浏览器扩展。

pip install bibguard        # Python
npx bibguard paper.bib      # Node.js

核心能力

• 幽灵 DOI / arXiv ID 检测 -- 格式正确但不存在 = 最强幻觉信号
• Kill-shot 逻辑 -- 幽灵 ID 不会被相似论文的搜索结果覆盖
• @misc 类型感知 -- 非论文条目（新闻、文档等）不会误报为幻觉
• TeX 交叉审计 -- 检测 \cite{key} 在 .bib 中无定义的条目
• 自动修复 -- 补全缺失的 DOI 和 eprint 字段

基准测试

测试集	幻觉检出	真论文假阳 (FAIL)	真论文 OK 率
Golden (58 条)	100% (14/14)	0%	--
大规模 (200 条)	100% (50/50)	0%	86%

TypeScript 版零依赖，可直接在浏览器运行（5 个 API 全支持 CORS）。支持 Claude Code / Codex / Cursor。

完整版请使用 IntegriRef（L0-L4，含语义 NLI 93.5%、贝叶斯风险评分）。

支持项目

如果 bibguard 对你有帮助，欢迎在 GitHub 上点个 Star ⭐，帮助更多人发现这个工具。有建议或问题？欢迎提 Issue。

路线图

• 并行验证 — 已在 v0.3.1 实现：--workers N 线程池并行验证
• 异步 I/O — 从 requests 迁移到 asyncio + aiohttp，进一步加速 500+ 条目文件
• 批量 API 查询 — 利用 Crossref、Semantic Scholar、OpenAlex 批量接口减少逐条开销
• 本地缓存 — 跨运行缓存已验证条目
• 撤稿论文检测 — 对接 Retraction Watch 数据库