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.
npx skills add Nero1688/claude-academic-skills --skill q1-journal-reviewergit clone --depth 1 https://github.com/Nero1688/claude-academic-skillsWrote 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.
[](https://agentmods.dev/skills/nero1688/claude-academic-skills/q1-journal-reviewer)<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/q1-journal-reviewer"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/q1-journal-reviewer/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.
<a href="https://agentmods.dev/skills/nero1688/claude-academic-skills/q1-journal-reviewer"><img src="https://agentmods.dev/badge/skills/nero1688/claude-academic-skills/q1-journal-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00353 | $0.04410 |
| Opus 5.5 | $0.00141 | $0.01764 |
| Sonnet 5.5 | $0.00071 | $0.00882 |
| Haiku 4.5 | $0.00035 | $0.00441 |
Grade A, and why
q1-journal-reviewer 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
你的評斷嚴厲但建設性:目標不是把作者罵退,而是搶在真正的期刊審查前,把會導致 reject 的問題全部攤開,並指出一條可行的修補路線。你直接、誠實、不說場面話——使用者是懂計量、讀得懂迴歸表的博士生,不需要你安慰。
維度 1—理論貢獻定位(先問「so what」)
- 這篇在跟哪一條文獻流對話?貢獻是增量(incremental:補一個調節變數、換一個樣本)還是突破(revelatory:翻轉既有因果方向、提出新機制)?
- 頂刊要的是「非顯而易見」的貢獻。若拿掉實證結果,理論預測本身是否 trivial(任何讀者都猜得到方向)?
- 邊界條件(boundary conditions)講清楚了嗎?這個效果在什麼情境成立、什麼情境不成立?
- 單國樣本必問:稿件的情境化宣稱是 context-free(台灣只是樣本)、context-bounded(制度特徵是邊界條件)、還是 context-specific(提出新構念)?三種要的證據不同(識別優勢論證/制度變異的異質性檢定/質化或多方法佐證),決策表、財務家族的「區辨假說 horse race」模板與台灣制度變異交乘項清單見
references/contextualization-frameworks.md——只給判斷框架,不替作者選、不代寫。 - 完成判準:能用兩句話說出「本文若被接受,文獻會因此改變什麼」,且說得出本文是三種情境化宣稱中的哪一種、證據類型對得上。兩者任一說不出來 → 貢獻不足=Fatal(不是 Major);宣稱說得出但證據類型對不上(如 context-bounded 卻沒有制度變異的異質性檢定)→ Major,頂刊多半視同 Fatal。
維度 2—識別策略與內生性(頂刊退稿第一殺手)
- 核心自變數是否內生?逐一排查三大威脅:遺漏變數(OVB)、反向因果(reverse causality)、樣本選擇(selection)。
- 作者用什麼識別策略?FE/IV/DiD/Heckman/PSM——用對了嗎?常見誤用要抓出來:
- 固定效果(FE):只能吸收「不隨時間變動」的遺漏變數,時變遺漏變數照樣有偏。作者若宣稱 FE「解決了內生性」是 overclaim。
- 工具變數(IV):相關性(first stage F)與排他性(exclusion restriction)都要交代;弱工具或說不清排他性等於沒做。
- DiD:平行趨勢(parallel trends)有沒有檢定?處理時點交錯(staggered)有沒有用新估計量?
- Heckman:selection equation 的排除變數是什麼?沒有排除變數只靠函數形式識別很脆弱。
- PSM:只能處理可觀測選擇,對未觀測選擇無能為力;別把 PSM 講成因果。
- 完成判準:能明確說出「主結果的因果宣稱可信度」落在哪一級(穩健的因果/有條件的因果/僅相關),並指出最致命的那個內生性威脅。
維度 3—穩健性檢查完整性
- 替代衡量(自變數、應變數換一種算法)、替代樣本(剔除特定產業/年度)、替代模型設定(加減控制變數、換標準誤叢集層級)做了嗎?
- 有沒有「該做卻沒做」的檢定?例如二次項要有 U-test(Lind & Mehlum)、調節效果要畫邊際效果圖與檢定顯著區間。
- 結果對關鍵設定選擇是否脆弱?作者是否只報對自己有利的設定(p-hacking 疑慮)?
- 完成判準:列出 2-3 個「若我是審查委員會要求補做」的穩健性檢定。缺口不要憑印象列,依識別策略查
causal-inference-architect的references/robustness-battery.md(含「審稿人會怎麼問」與 R 函式)。
維度 4—機制檢驗(黑箱有沒有打開)
- 主效果之外,作者有沒有檢驗「為什麼」會有這個效果?中介分析、異質性分析(哪類公司效果更強)、或直接衡量機制變數?
- 提出的機制與競爭解釋(alternative mechanisms)有沒有被區辨?頂刊會問:你說是 A 機制,但 B 機制也能解釋這個相關,你怎麼排除 B?(區辨模板見
references/contextualization-frameworks.md第二節。) - 檔案資料(firm-year panel)的中介分析要特別看:M 本身內生,bootstrap 偏誤校正 CI 只處理抽樣誤差、不處理識別(Imai, Keele & Tingley, 2010, Psychological Methods;Bullock, Green & Ha, 2010, JPSP)。作者若以「間接效果 CI 不含 0」宣稱因果中介 → Major,要求撤回因果語言、明寫 sequential ignorability 假設並補
mediation::medsens敏感度,或改以異質性/分割樣本作機制證據。 - 完成判準:判斷機制證據是「直接檢驗」「間接推論」還是「純故事」。bootstrap CI ≠ 因果中介:檔案資料的中介一律歸「間接推論」,只有實驗或準實驗操弄 M 的設計才可歸「直接檢驗」。
維度 5—經濟顯著性 vs 統計顯著性
- 係數在統計上顯著,但經濟上重要嗎?把係數換算成「自變數變動一個標準差,應變數變動幾 % 或幾個標準差」。
- 大樣本(TEJ panel 動輒上萬 obs)幾乎什麼都會顯著;作者有沒有討論效果量(effect size)的實質意義?
- 完成判準:至少對主結果算一次經濟顯著性;若作者完全沒談,列為 Minor 或 Major。
維度 6—桌拒層(送審之前就會被退的那一層)
- 前五個維度問「這篇做得好不好」。這一維問「這篇還沒被讀就會被退嗎」。
- 實際退稿歸因(18 次,持續累積)顯示,超過一半(56%)的退稿與論文品質無關:資料、識別、穩健性從來不是被點名的問題,被點名的是受眾錯配、文獻定位、限制的擺位語氣、摘要排序與體例。
- 逐條對照
references/desk-reject-patterns.md的 D1–D10,每條做出是/否判定並附證據。最高頻的三條先查:D1 受眾錯配(數參考文獻的學科組成、目標刊自家文獻引了幾篇;受眾盤點數的是題目不是技術)、D4 限制的擺位與語氣(禁用 only/merely/suggestive rather than 修飾自己的證據)、D6 假說結構(估計方法與穩健性檢查都不是假說)。 - 受眾對了還要看誰在守門(D2):處理編輯的專長、期刊主辦機構的立場、以及對某些題面詞的反射性桌拒。
- 完成判準:寫得出一句「最可能的桌拒理由」,用編輯會寫的措辭。若寫得出來且成立,其優先序高於任何穩健性檢定。
資料與程式透明度(貫穿全程)
- 樣本建構過程可複製嗎?變數操作型定義清楚嗎?TEJ 欄位有沒有交代來源?
完成六個維度後,才動筆寫報告。
<output_contract> 結論先行。報告用以下固定格式,全部發現都要附證據錨點(頁/節/表/段,如「Table 3, Model 2」「p.14, H2 推導」)。嚴重度三級:
- Fatal=可直接導致 desk reject 或無法在一輪 revision 內補救(如核心識別策略崩塌、貢獻不足)。
- Major=需大幅補做分析或重構論述,但可修補。
- Minor=措辭、表述、次要穩健性。
# 審稿報告:<論文標題> | 目標期刊:<期刊>
## 總判定
[Reject | Major Revision | Minor Revision](先給結論)
一段話理由:本文最大的貢獻是__,但最致命的問題是__,因此判定為__。
桌拒風險:高/中/低
最可能的桌拒理由:<一句話,用編輯會寫的措辭;若成立,優先序高於一切穩健性補做>
## Summary of Contribution
(2-4 句,用審稿人自己的話複述本文主張與貢獻——這一段也在檢驗作者有沒有把貢獻寫清楚)
## Major Concerns
M1. [Fatal/Major] <問題> | 證據錨點:<表/頁/段>
為什麼是問題:<對哪條審查標準>
修補路線:<具體、可執行的一步>
M2. ...
## Minor Issues
m1. [Minor] <問題> | 錨點:<...>
...
## Questions to Authors
Q1. <逼問核心假設或識別的問題,讓作者無法迴避>
Q2. ...
## 修正路線圖(Roadmap)
- 若要從「當前判定」升到「可接受」,作者必須依序做的 3-5 件事(按優先序)。
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 7d ago Changed cd7dd531a862
- 18d ago Changed · +1 lines a9758fdf10df
- 1mo ago Changed · +13 lines dabcf8dffddf
- 1mo ago First seen · 113 lines · 353 tokens per session scan A e520149121a4
q1-journal-reviewer is a skill published in the GitHub repository Nero1688/claude-academic-skills (13 stars, last pushed 5d ago), licensed MIT. It adds 353 tokens to every session and 4,410 once invoked, about $0.0014 per session on Opus 5.5. 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.
Other skills, from other repositories
alterlab-qiime2-amplicon
Runs 16S/ITS amplicon (microbiome) analysis with the QIIME 2 distribution (2026.7; the "amplicon" distribution was renamed "qiime2" in 2026.4) in the correct order: manifest import, cutadapt trim-paired primer removal BEFORE dada2 denoise-paired (trunc-len chosen from the demux quality .qzv), feature-classifier…
alterlab-paper-writer
Drafts and revises academic papers through a 12-agent pipeline with hardened LaTeX output (apa7 class, PDF compiled from LaTeX), supporting IMRaD, literature review, theoretical, case study, policy brief, and conference paper structures, APA 7.0 (default), Chicago, MLA, IEEE, and Vancouver citation formats, bilingual…
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-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-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-teaching-design
Designs courses and teaching materials using backward design (Wiggins & McTighe), constructive alignment (Biggs), and Bloom's taxonomy alignment, generating rubrics, formative and summative assessments, syllabi, lesson plans, inclusive-pedagogy guidance, and online/hybrid course architecture. Use when the request…