Imbad0202 / academic-research-skills

查看原文 / View Original 编辑 / Edit

Imbad0202 / academic-research-skills

Academic Research Skills for Claude Code 繁體中文版

A comprehensive suite of Claude Code skills for academic research, covering the full pipeline from research to publication. Install in 30 seconds (Claude Code CLI / VS Code / JetBrains, v3.7.0+): /plugin marketplace add Imbad0202/academic-research-skills /plugin install academic-research-skills Then try /ars-plan to walk through your paper structure via Socratic dialogue, or jump to Quick install for prerequisites and the traditional symlink flow.

這是一套為 Claude Code 設計的學術研究技能套件，涵蓋了從研究到發表的完整流程。您可以在 30 秒內完成安裝（適用於 Claude Code CLI / VS Code / JetBrains，版本 v3.7.0+）： /plugin marketplace add Imbad0202/academic-research-skills /plugin install academic-research-skills 安裝後，請嘗試輸入 /ars-plan，透過蘇格拉底式對話來規劃您的論文架構，或直接跳至「快速安裝」查看前置需求與傳統符號連結（symlink）流程。

AI is your copilot, not the pilot. This tool won’t write your paper for you. It handles the grunt work — hunting down references, formatting citations, verifying data, checking logical consistency — so you can focus on the parts that actually require your brain: defining the question, choosing the method, interpreting what the data means, and writing the sentence after “I argue that.”

AI 是您的副駕駛，而非機長。此工具不會代您撰寫論文。它負責處理繁瑣的工作——搜尋參考文獻、格式化引用、驗證數據、檢查邏輯一致性——讓您可以專注於真正需要大腦思考的部分：定義問題、選擇方法、詮釋數據意義，以及撰寫「我主張（I argue that）」之後的論述。

Unlike a humanizer, this tool doesn’t help you hide the fact that you used AI. It helps you write better. Style Calibration learns your voice from past work. Writing Quality Check catches the patterns that make prose feel machine-generated. The goal is quality, not cheating.

與「去 AI 化（humanizer）」工具不同，本工具不會協助您隱瞞使用 AI 的事實，而是幫助您寫得更好。「風格校準（Style Calibration）」會從您過去的作品中學習您的語氣；「寫作品質檢查（Writing Quality Check）」則能捕捉那些讓文章顯得像機器生成的模式。我們的目標是品質，而非作弊。

Why human-in-the-loop, not full automation? Lu et al. (2026, Nature 651:914-919) built The AI Scientist — the first fully autonomous AI research system to publish a paper through blind peer review at a top-tier ML venue (ICLR 2025 workshop, score 6.33/10 vs workshop average 4.87). Their Limitations section enumerates the failure modes that any fully-autonomous AI research pipeline inherits: implementation bugs, hallucinated results, shortcut reliance, bug-as-insight reframing, methodology fabrication, frame-lock, citation hallucinations. ARS is built on the premise that a human researcher augmented by AI avoids these failure modes better than either alone.

為什麼強調「人在迴路（human-in-the-loop）」而非全自動化？Lu 等人（2026, Nature 651:914-919）開發了「AI 科學家（The AI Scientist）」，這是第一個透過頂尖機器學習會議（ICLR 2025 工作坊，得分 6.33/10，高於平均 4.87）盲審並發表論文的全自動 AI 研究系統。他們的「限制」章節列舉了任何全自動 AI 研究流程都會繼承的失敗模式：實作錯誤、幻覺結果、依賴捷徑、將錯誤誤解為洞見、捏造方法論、框架鎖定以及引用幻覺。ARS 的核心前提是：由 AI 輔助的人類研究者，比單獨的人類或單獨的 AI 更能避免這些失敗模式。

Stage 2.5 and Stage 4.5 integrity gates run a 7-mode blocking checklist (see academic-pipeline/references/ai_research_failure_modes.md); the reviewer offers an opt-in calibration mode that measures its own FNR/FPR against a user-supplied gold set.

第 2.5 階段與第 4.5 階段的完整性閘道會執行一套 7 模式的阻斷檢查清單（參見 academic-pipeline/references/ai_research_failure_modes.md）；審閱者提供了一種可選的校準模式，能針對使用者提供的黃金標準集（gold set）來衡量自身的偽陰性率（FNR）與偽陽性率（FPR）。

Zhao et al. (2026-05) audited 111M references across 2.5M papers on arXiv, bioRxiv, SSRN, and PMC. Their conservative estimate is 146,932 hallucinated citations for 2025 alone, with an observed mid-2024 inflection; for the bioRxiv-to-PMC pairing they report 85.3% preprint-to-published persistence. The paper describes “real citations deployed to support claims the cited references do not actually make” as an open challenge. ARS v3.7.1 added trust-chain frontmatter for source provenance; v3.7.3 added locator infrastructure (three-layer citation anchors) for future claim-level audits and surfaces advisory risk signals at cite time (ARS labels the claim-faithfulness gap internally as “L3”; this is ARS terminology, not the paper’s).

Zhao 等人（2026-05）審計了 arXiv、bioRxiv、SSRN 和 PMC 上 250 萬篇論文中的 1.11 億條參考文獻。保守估計，僅 2025 年就有 146,932 條引用屬於幻覺，且在 2024 年年中出現了轉折點；在 bioRxiv 到 PMC 的配對中，他們報告了 85.3% 的預印本轉為正式發表的一致性。該論文將「引用真實文獻但用來支持該文獻並未提出的主張」描述為一項待解決的挑戰。ARS v3.7.1 新增了用於來源溯源的信任鏈前言；v3.7.3 新增了定位器基礎設施（三層引用錨點），以便未來進行主張層級的審計，並在引用時顯示風險警示訊號（ARS 將這種主張忠實度差距內部標記為「L3」；這是 ARS 的術語，非論文原用語）。

v3.7.x is motivated by Zhao et al.’s corpus-scale findings; corpus-scale evaluation of ARS itself remains future work. v3.8 closes the second half of the L3 gap. v3.7.3 made every citation carry a locator anchor; v3.8 adds an opt-in audit pass (ARS_CLAIM_AUDIT=1) that fetches the cited source against each anchor and judges whether the claim is actually supported. Five new HIGH-WARN classes (claim-not-supported, negative-constraint-violation, fabricated-reference, anchorless, constraint-violation-uncited) gate-refuse output through the formatter terminal hard gate.

v3.7.x 的開發動機源於 Zhao 等人的語料庫規模研究發現；ARS 本身的語料庫規模評估仍是未來的工作。v3.8 補足了 L3 差距的後半部分。v3.7.3 讓每條引用都帶有定位錨點；v3.8 新增了可選的審計流程（ARS_CLAIM_AUDIT=1），該流程會根據每個錨點抓取引用的來源，並判斷該主張是否確實得到支持。五種新的「高風險警告（HIGH-WARN）」類別（主張未獲支持、違反負面約束、捏造參考文獻、無錨點、違反約束且未引用）會透過格式化終端機的硬性閘道拒絕輸出。

Calibration is shipped as a 20-tuple gold set with FNR<0.15 + FPR<0.10 acceptance thresholds; ramp-on plan is deferred to post-calibration evidence per v3.8 spec §5. v3.3 was inspired by PaperOrchestra (Song, Song, Pfister & Yoon, 2026, Google): Semantic Scholar API verification, anti-leakage protocol, VLM figure verification, and score trajectory tracking.

校準功能隨附一組 20 個項目的黃金標準集，接受閾值為 FNR < 0.15 且 FPR < 0.10；根據 v3.8 規範第 5 節，導入計畫將延後至校準後的證據確認後進行。v3.3 的靈感來自 PaperOrchestra（Song, Song, Pfister & Yoon, 2026, Google）：包含 Semantic Scholar API 驗證、防洩漏協定、VLM 圖表驗證以及評分軌跡追蹤。

Architecture & pipeline

👉 docs/ARCHITECTURE.md — the full pipeline view: flow diagram, stage-by-stage matrix, data-access flow, skill dependency graph, quality gates, and mode list. The architecture doc supersedes the sprawling pipeline description that used to live here. Everything about what runs in which stage now lives in one place.

👉 docs/ARCHITECTURE.md — 完整的流程視圖：包含流程圖、階段矩陣、數據存取流、技能依賴圖、品質閘道以及模式列表。此架構文件取代了過去散落在各處的冗長流程描述。現在，關於每個階段執行內容的所有資訊都集中在同一個地方。

Quick install

Prerequisites

  • Claude Code (latest; plugin packaging requires recent versions)
  • ANTHROPIC_API_KEY exported, or set on first claude run
  • Optional: Pandoc for DOCX, tectonic + Source Han Serif TC for APA 7.0 PDF (Markdown output works without either)

Plugin install (v3.7.0+, recommended): /plugin marketplace add Imbad0202/academic-research-skills /plugin install academic-research-skills

Verify it works: run /ars-plan and describe a paper you’re working on — ARS will start a Socratic dialogue to map out chapter structure. For a single-shot test instead, try /ars-lit-review "your topic".

👉 docs/SETUP.md — full guide: install Claude Code, set up API keys, optional Pandoc/tectonic for DOCX/PDF, cross-model verification (ARS_CROSS_MODEL), and five installation methods (Plugin, project skills, global skills, claude.ai Project, repo-cloned).

Using Codex CLI? Install the sibling distribution instead: Imbad0202/academic-research-skills-codex — same workflow content, Codex-native packaging as a single $academic-research-suite skill with ars-* aliases.

快速安裝

前置需求

  • Claude Code（最新版；插件封裝需要較新版本）
  • 已匯出 ANTHROPIC_API_KEY，或在首次執行 claude 時設定
  • 選用：用於 DOCX 的 Pandoc，用於 APA 7.0 PDF 的 tectonic + 思源宋體 TC（Markdown 輸出無需上述工具即可運作）

插件安裝（v3.7.0+，推薦）： /plugin marketplace add Imbad0202/academic-research-skills /plugin install academic-research-skills

驗證是否運作： 執行 /ars-plan 並描述您正在進行的論文——ARS 將啟動蘇格拉底式對話來規劃章節架構。若要進行單次測試，請嘗試 /ars-lit-review "您的主題"。

👉 docs/SETUP.md — 完整指南：安裝 Claude Code、設定 API 金鑰、選用的 DOCX/PDF 工具（Pandoc/tectonic）、跨模型驗證（ARS_CROSS_MODEL），以及五種安裝方法（插件、專案技能、全域技能、claude.ai Project、repo-cloned）。

使用 Codex CLI？請改為安裝姊妹版本：Imbad0202/academic-research-skills-codex — 內容與工作流相同，採用 Codex 原生封裝為單一 $academic-research-suite 技能，並提供 ars-* 別名。

Performance & cost

👉 docs/PERFORMANCE.md — per-mode token budgets, full-pipeline estimate (~$4–6 for a 15k-word paper), and recommended Claude Code settings (Skip Permissions; Agent Team optional).

👉 docs/PERFORMANCE.md — 各模式的 Token 預算、完整流程預估費用（1.5 萬字論文約需 4–6 美元），以及推薦的 Claude Code 設定（跳過權限檢查；Agent Team 為選用）。

Guides & articles

  • Academic Writing Shouldn’t Be a Solo Act — full pipeline walkthrough (English)
  • 學術寫作不該是一個人的事：一套開源 AI 協作工具如何改變研究者的工作流 — 完整使用指南（繁體中文）

Features at a glance

  • Deep Research — 13-agent research team with Socratic guided mode, PRISMA systematic review, intent detection, dialogue health monitoring, optional cross-model DA, Semantic Scholar API verification.
  • Academic Paper — 12-agent paper writing with Style Calibration, Writing Quality Check, LaTeX hardening, visualization, revision coaching, citation conversion, anti-leakage protocol, and VLM figure verification.
  • Academic Paper Reviewer — 7-agent multi-perspective peer review with…

功能概覽

  • 深度研究 (Deep Research) — 13 個代理人的研究團隊，具備蘇格拉底引導模式、PRISMA 系統性回顧、意圖偵測、對話健康監控、選用的跨模型 DA 以及 Semantic Scholar API 驗證。
  • 學術論文 (Academic Paper) — 12 個代理人的論文寫作團隊，具備風格校準、寫作品質檢查、LaTeX 強固、視覺化、修訂指導、引用轉換、防洩漏協定以及 VLM 圖表驗證。
  • 學術論文審閱者 (Academic Paper Reviewer) — 7 個代理人的多視角同儕審閱，具備…
← 返回新闻列表 / Back to news list