citation-verifier

citation-verifier is a skill for Claude Code, Codex from Nero1688/claude-academic-skills. It costs 337 tokens per session (2,484 once invoked), scanned A, original, MIT.

A tool for checking whether in-text citations match the reference list and whether references follow APA 7 formatting. It also flags signs that a citation may be incorrect or fabricated, but cannot confirm a source exists without external verification.

In plain words
What is it for?
Use it to audit academic papers, reconcile citations with reference lists, check APA 7 entries, inspect DOI and page details, and prepare a list of items needing online verification.
Why use it?
It helps find missing or unused references, formatting mistakes, and suspicious details without pretending that an offline check proves a source is real.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Claude Code.

Good fit Use it to audit academic papers, reconcile citations with reference lists, check APA 7 entries, inspect DOI and page details, and prepare a list of items needing online verification.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/nero1688/claude-academic-skills/citation-verifier
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Nero1688/claude-academic-skills --skill citation-verifier
Clone the repo
git clone --depth 1 https://github.com/Nero1688/claude-academic-skills

Made for: Claude Code, Codex.

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 citation-verifier

README.md
[![agentmods](https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/citation-verifier/github.svg)](https://agentmods.dev/skills/nero1688/claude-academic-skills/citation-verifier)
Your own site
<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/citation-verifier"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/citation-verifier/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 citation-verifier

Your own site · 80×15
<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/citation-verifier"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/citation-verifier.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 337 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,484 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 $0.00337 $0.02484
Opus 5 $0.00169 $0.01242
Sonnet 5 $0.00067 $0.00497
Haiku 4.5 $0.00034 $0.00248

Measured 12d ago against content hash 422d15ba0241, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

citation-verifier 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 12d 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.

skills/citation-verifier/SKILL.md · 86 lines

What it actually says

這條能力邊界是本 skill 的核心。原始版本承諾「驗證文獻真實存在」卻沒有驗證工具,會逼模型假裝查過——這正是要杜絕的。你寧可說「疑似、需工具驗證」,也不假造查證結果。

階段一:雙向對帳(可離線完整完成,最高信心) 這是本 skill 最扎實的部分——不需上網,純比對內文與清單。

  • 孤兒引用(內文有、清單無):揪出所有正文出現(如 Smith et al., 2023)卻未列入參考文獻清單的項目。
  • 冗餘文獻(清單有、內文無):揪出列在清單卻正文從未提及的項目。
  • 對帳時注意 et al. 展開、同作者同年 a/b 標註(2023a vs 2023b)、中英文姓名對應。
  • 完成判準:給出兩張對不上的清單,逐筆列「內文位置 vs 清單狀態」。這一階段可下定論,不必標「疑似」。

階段二:APA 7 格式診斷(可離線,依規則判斷) 逐筆檢查參考文獻清單的 APA 7 合規:

  • 期刊名與卷號應斜體、篇名句首大寫(sentence case)、DOI 以 https://doi.org/ 格式呈現。
  • 作者列表格式(& 用法、7 位以上作者省略規則)、年份括號位置、頁碼範圍格式。
  • et al. 內文用法(3 位以上作者第一次即用 et al.)。
  • 完成判準:逐筆標出不合規處與正確寫法。格式對錯可離線定論。

階段三:幻覺風險訊號標記(只標風險,不下真偽結論) 以下訊號提高「可能是 AI 捏造或誤植」的機率,但不能單獨證明文獻是假的——一律標「⚠ 疑似,需工具驗證」:

  • DOI 格式錯:不以 10.xxxx/ 開頭、格式殘缺。(格式錯可離線判定;但格式對 ≠ 真實存在。)
  • 頁碼範圍異常:起頁 > 迄頁、範圍過長(如 1–500)、與該期刊常見篇幅明顯不符。
  • 作者-年份-期刊組合可疑:期刊當年尚未創刊/已停刊、作者領域與期刊主題明顯錯配、年份晚於當前日期(未來出版)。
  • 標題過度泛化:如「A Comprehensive Survey of...」配上具體實證宣稱,AI 生成常見樣態。
  • chimeric 疑慮:真實感很強的標題配上組合突兀的作者——但是否真為拼接,離線無法確認,只能標「疑似 chimeric,需上網比對作者是否真發表過此文」。
  • 完成判準:每個風險訊號都寫「為什麼可疑」+「需要什麼工具/資料才能定論」。

<output_contract> 結論先行。用表格+列點。嚴重度三級:

  • Fatal=幾乎確定有問題且會傷學術聲譽(孤兒引用整段無來源、DOI 明顯偽造格式)。
  • Major=高風險需優先查證(多個幻覺訊號疊加、疑似 chimeric)。
  • Minor=格式不合規、單一弱訊號。
# 引用稽核報告(無網/內部一致性+風險評估)
## 能力邊界聲明
本報告在無網路環境完成,未連任何文獻資料庫,故【無法確認任何文獻真實存在】。
標「⚠ 需工具驗證」者請以 check-citations(有網)查驗。

## 一、雙向對帳(可下定論)
孤兒引用(內文有、清單無):<列表+內文位置>
冗餘文獻(清單有、內文無):<列表>

## 二、APA 7 格式診斷(可下定論)
[Minor] <第 N 筆> 問題:<...> | 正確寫法:<...>

## 三、幻覺風險訊號(僅標疑似,需工具驗證)
[Major] ⚠ <第 N 筆> 訊號:<DOI格式錯/頁碼異常/組合可疑...> | 為何可疑:<...> | 定論需要:<上網比對 X>

## 建議下一步
有網路時,用 check-citations 對接 CrossRef/Semantic Scholar/OpenAlex 查驗上列「⚠」項目
(Claude Code 環境須加 anthropic-skills: 前綴)。裸名引用可直接說「跑 check-citations」。

</output_contract>

輸出節錄:

  • 一、雙向對帳 [Fatal]:內文「Chen (2019)」為孤兒引用,參考文獻清單無此筆——內文位置在第 3 段討論代理問題處。請補列或刪除引用。
  • 二、APA 7 [Minor]:Villalonga & Amit (2006) 該筆篇名、卷期格式符合 APA 7;期刊名 Journal of Financial Economics 應斜體(純文字稿無法判斷是否已斜體,請作者確認)。
  • 三、風險訊號 [Major] ⚠:DOI「JFE.2006.0385」格式錯誤,未以 10. 開頭,非合法 DOI——可離線判定格式錯;但「這篇文獻本身是否真實」離線無法確認。此筆為公司治理/家族企業高被引經典,作者-年份-期刊組合合理,幻覺機率低,但仍需上網定論。定論需要:用 check-citations 查 CrossRef 取得正確 DOI(10.1016/j.jfineco.2004.12.005 之類)並確認存在。
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. 12d ago First seen · 86 lines · 337 tokens per session scan A 422d15ba0241

Subscribe to this mod's changes

citation-verifier is a skill published in the GitHub repository Nero1688/claude-academic-skills (6 stars, last pushed 9d ago), licensed MIT. It adds 337 tokens to every session and 2,484 once invoked, about $0.0017 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 2026-08-31.

Related

Other skills, from other repositories

alterlab-deep-research

Runs a 13-agent deep research pipeline for rigorous academic work on any topic across 7 modes (full research, quick brief, paper review, lit-review, fact-check, Socratic guided research dialogue, and systematic review with optional meta-analysis), covering research-question formulation, Socratic mentoring, methodology…

AlterLab-IEU/AlterLab-Academic-Skills · 239 tokens

alterlab-paper-writer

Drafts and revises academic papers through a 12-agent pipeline with hardened LaTeX output (apa7 document class, justified text, table column-width formula, centered bilingual abstracts, standardized font stack, PDF compiled from LaTeX), supporting IMRaD, literature review, theoretical, case study, policy brief, and…

AlterLab-IEU/AlterLab-Academic-Skills · 276 tokens

alterlab-research-pipeline

Orchestrates the full academic research pipeline (research, write, integrity check, review, revise, re-review, re-revise, final integrity check, finalize), coordinating alterlab-deep-research, alterlab-paper-writer, and alterlab-paper-reviewer into a seamless 10-stage workflow with mandatory integrity verification…

AlterLab-IEU/AlterLab-Academic-Skills · 126 tokens

alterlab-imaging-data-commons

Query and download public cancer imaging data from the NCI Imaging Data Commons (IDC) using the idc-index Python package, filtering by metadata, visualizing in-browser, and checking licenses, with no authentication required. Use when obtaining large-scale radiology (CT, MR, PET) or digital pathology DICOM datasets for…

AlterLab-IEU/AlterLab-Academic-Skills · 90 tokens

alterlab-pyhealth

Develops, tests, and deploys clinical machine learning models with the PyHealth healthcare AI toolkit. Use when working with electronic health records (EHR), clinical prediction tasks (mortality, readmission, drug recommendation), medical coding systems (ICD, NDC, ATC), physiological signals (EEG, ECG), healthcare…

AlterLab-IEU/AlterLab-Academic-Skills · 117 tokens

alterlab-paper-reviewer

Simulates a full multi-reviewer journal review PANEL — 5 personas (Editor-in-Chief + 3 peer reviewers + a Devil's Advocate) debate a manuscript and produce a consensus Editorial Decision (accept/minor/major/reject) plus a prioritized Revision Roadmap. Modes: full, re-review (verify revisions addressed prior comments)…

AlterLab-IEU/AlterLab-Academic-Skills · 166 tokens