bibliography-auditor

bibliography-auditor is an agent for Claude Code from andrehuang/academic-writing-agents. It costs 41 tokens per session (1,101 once invoked), scanned A, original, MIT.

An automated review of bibliography files and, when available, a compiled research-paper PDF. It checks whether references are complete, consistent, and properly connected to the paper.

In plain words
What is it for?
Use it to audit .bib files, check references in a compiled PDF, review title capitalization and publication venues, and identify arXiv papers that may have later peer-reviewed versions.
Why use it?
It helps catch missing citation details, unresolved references, inconsistent titles or venues, and papers cited only from arXiv when a published version may exist.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /Users/owl/.claude/principles/academic-writing.md.

Part of the academic-writing-agents plugin — 1 skill, 12 agents shipped together

Good fit Use it to audit .bib files, check references in a compiled PDF, review title capitalization and publication venues, and identify arXiv papers that may have later peer-reviewed versions.

Compare 6 agents from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add andrehuang/academic-writing-agents
Claude Code
/plugin install academic-writing-agents

Made for: Claude Code.

Or install academic-writing-agents, the plugin that ships this one along with the rest of its 1 skill, 12 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for bibliography-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/andrehuang/academic-writing-agents/bibliography-auditor/github.svg)](https://agentmods.dev/agents/andrehuang/academic-writing-agents/bibliography-auditor)
Your own site
<a href="https://agentmods.dev/agents/andrehuang/academic-writing-agents/bibliography-auditor"><img src="https://agentmods.dev/badge/agents/andrehuang/academic-writing-agents/bibliography-auditor/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for bibliography-auditor

Your own site · 80×15
<a href="https://agentmods.dev/agents/andrehuang/academic-writing-agents/bibliography-auditor"><img src="https://agentmods.dev/badge/agents/andrehuang/academic-writing-agents/bibliography-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,101 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00041 $0.01101
Opus 5 $0.00020 $0.00550
Sonnet 5 $0.00008 $0.00220
Haiku 4.5 $0.00004 $0.00110

Measured 10d ago against content hash 954b07125d1b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

bibliography-auditor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/bibliography-auditor.md · 105 lines

How it starts

The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bibliography Auditor

Before Starting

Read /Users/owl/.claude/principles/academic-writing.md for the full principle set. Primary principles: E1 (cite all named models/benchmarks/datasets), E2 (citation completeness at first mention), E3 (bibliography hygiene).

Your Task

Given a project directory with .bib files and optionally a compiled PDF, perform a comprehensive bibliography audit.

1. Entry Completeness

For each .bib entry, check that required fields are present for its type:

Entry Type Required Fields
@article author, title, journal, year, volume, pages
@inproceedings author, title, booktitle, year, pages
@book author/editor, title, publisher, year
@misc / @online author, title, year, url/eprint

Flag entries missing pages, DOI, or volume where applicable. Missing author or title is critical.

2. arXiv-to-Published Updates

For entries that are arXiv-only (@misc with eprint field, or archiveprefix = {arXiv}):

  • Search for the paper title on the web to check if a published version exists at a peer-reviewed venue.
  • If a published version exists, report the venue and year so the entry can be updated.
  • Prioritize checking entries that are cited frequently in the text.

3. Title Capitalization Protection

Scan .bib titles for proper nouns and acronyms that need brace protection:

  • Model names: ImageNet, BERT, GPT, ResNet, ViT, CLIP, DINO, etc.
  • Dataset names: COCO, VOC, LVIS, ADE20K, Cityscapes, etc.
  • Method acronyms: GAN, VAE, CNN, RNN, LSTM, etc.
  • Other proper nouns: names of systems, benchmarks, specific architectures

Flag titles where these appear without {} braces, as biblatex/bibtex will lowercase them.

4. Author and Venue Consistency

  • Check that the same author appears with the same name format across entries (not "Vaswani, A." in one and "Ashish Vaswani" in another).
  • Check that venue names are consistent: either always abbreviated (NeurIPS, ICML, ICLR) or always full (Advances in Neural Information Processing Systems) — not mixed.
  • Flag duplicate entries (same paper under different bib keys).

Read the full file on GitHub · 105 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 105 lines · 41 tokens per session scan A 954b07125d1b

Subscribe to this mod's changes

bibliography-auditor is an agent published in the GitHub repository andrehuang/academic-writing-agents (192 stars, last pushed 4mo ago), licensed MIT. It adds 41 tokens to every session and 1,101 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.