Academic Research Skills - Open Source AI Agent for Academic Research Skills - AiBoss

What are Academic Research Skills?

Academic Research Skills is an open-source academic skills plugin developed by Cheng-I Wu for Claude Code, covering the entire process from topic selection and research, paper writing, peer review, to final revisions. The plugin utilizes collaboration from 39+ specialized agents to provide functions such as citation source verification, Socratic research guidance, anti-illusion verification, and experimental source registration. The core goal of Academic Research Skills is to help researchers transform fragmented information into complete papers that conform to academic standards.

Main functions of Academic Research Skills

In-depth researchThirteen intelligent agents collaborate to complete literature retrieval, PRISMA systematic reviews, and Socratic research guidance, helping you extract falsifiable research questions from fragmented information.
Thesis writing12 intelligent agents cover outlines, abstracts, reviews, revisions to the final draft, supporting style calibration and bilingual abstract generation.
Peer reviewSeven intelligent agents simulate journal reviewers plus a devil's advocate, providing multi-dimensional, evidence-anchored narrative evaluations of papers.
Assembly Line ArrangementThe 10-stage full-process orchestrator sets mandatory quality inspection gates at key nodes to ensure that each step is only advanced after user confirmation.
Citation tracing verificationEach citation is cross-verified using four indexes: Semantic Scholar, OpenAlex, Crossref, and arXiv, and three layers of anchor points are added to audit the fidelity of the claim.
Experimental traceability registrationRecord in Material Passport the reproducibility locks, negative results, and known limitations of external experiments, and audit the alignment between the paper claims and experimental evidence.
Anti-illusion and anti-flatteryBuilt-in devil's advocate concession threshold protocol and intent detection layer to prevent AI from converging too early or compromising without principle when you insist.
Cross-model validationIt supports calling the second model family to independently perform peer review and quality control, reducing the blind spots in systematic understanding of a single model.
Multi-format deliveryOne-click output to Markdown, DOCX, and LaTeX (APA 7.0 / IEEE / Chicago) and compilation to PDF.

How to use Academic Research Skills

Install: Execute in Claude Code /plugin marketplace add Imbad0202/academic-research-skills One-click installation.
Topic planning:run /ars-planThe research questions and chapter structure are organized through Socratic dialogue.
Literature reviewUsing Deep Research full or systematic-review Pattern retrieval and systematic literature review.
Experiment Registration(If applicable): Register the reproducibility locks and negative results of external experiments in Material Passport.
Thesis writing: Calling Academic Paper full In this model, 12 agents work in an assembly line to complete the outline and draft.
Mid-term quality inspectionThe Stage 2.5 mandatory property check gate cross-checks the existence and logical consistency of references.
Peer reviewUsing Academic Paper Reviewer full The mode is activated by a virtual peer review panel of 5 people to conduct multi-dimensional reviews.
Revised ResponseRevise according to the reviewers' comments and generate an R&R traceability matrix to respond to each item.
Final quality inspectionThe claim was re-audited in Stage 4.5 to align with experimental evidence, confirming zero regression.
Final draft deliverySelect APA/IEEE/Chicago format to output LaTeX/PDF and generate a 6-dimensional collaboration quality assessment report.

Core Advantages of Academic Research Skills

Full-process closed loopIt covers the entire academic pipeline from topic selection, research, writing, review, revision to submission, without the need to switch between multiple tools.
Human-computer collaborative designAdhering to the concept that "AI is the co-pilot," we require users to confirm at every key point to avoid academic misconduct caused by AI-assisted writing.
Citations are traceableCross-verification using four indexes (Semantic Scholar, OpenAlex, Crossref, and arXiv) and three layers of anchor auditing ensure claim fidelity and prevent illusory citations.
Auditable experimentsSupports external experimental traceability registration and automatically audits the alignment between paper claims and experimental evidence to prevent ablation experiments from being included in the main text when they were not performed.
Anti-flattery mechanismThe Devil's Advocate employs a concession threshold protocol to prevent AI from compromising without principle when users insist, thus ensuring the rigor of the research design.
Cross-model validationSupports independent peer review and quality control for the second model family, reducing the risk of systemic cognitive blind spots and framework lock-in for a single model.

The Academic Research Skills project address

GitHub repository：https://github.com/Imbad0202/academic-research-skills

Application scenarios of Academic Research Skills

Computer Empirical ResearchStarting from the fuzzy observation that "the model performs better", it is broken down into falsifiable hypotheses and ablation experimental designs. After being written in the IMRaD structure, the quality control gate intercepts the illusion of reference and statistical error.
Systematic Literature ReviewPerforms searches, inclusion/exclusion, risk bias assessment, and compliance report generation according to the PRISMA protocol, directly outputting review or meta-analysis papers that conform to academic standards.
Paper revisions and peer review responses: Analyze peer review comments to generate a revision roadmap, and use rebuttal-audit Verify each response to ensure it covers all questions and prevent omissions of key baseline comparisons.
Interdisciplinary topic explorationWhen the research direction is not yet clear, exploratory Socratic dialogue can prevent AI from converging prematurely and help extract theoretically valuable research questions from fragmented information.
Academic writing instructionThe course incorporates a 10-stage pipeline, allowing students' initial drafts to be reviewed by a virtual 5-person review panel, followed by classroom revision workshops based on structured review reports.