A human-in-the-loop Claude Code skill suite that supports academic research workflows such as reference hunting, citation formatting, and data verification. It requires the Anthropic API and Claude Code CLI, uses a mandatory integrity gate system to mitigate LLM failure modes, and prohibits commercial use under CC BY-NC 4.0.
Project overview
Addresses autonomous AI research pipeline failure modes (such as hallucinated results and citation fabrication) by enforcing mandatory, non-skippable integrity verification gates alongside a human-in-the-loop interaction model.
Researchers and educators who require a human-in-the-loop pipeline to manage reference hunting, citation formatting, and peer review.
Users seeking to mitigate autonomous AI hallucination risks through mandatory, non-skippable integrity verification gates.
Key capabilities
Provides a 7-agent multi-perspective peer review system with 0–100 quality rubrics, including an Editor-in-Chief, 3 dynamic reviewers, and a Devil's Advocate. It features a concession threshold protocol and a Revise & Resubmit traceability matrix.
Limitations and risks
The system never runs experiments; the scholar is responsible for running all experiments externally.
The project is licensed under CC BY-NC 4.0, strictly prohibiting commercial use.
If outputs are not properly audited despite mitigation, LLMs may still hallucinate citations.
Getting started
Setup is evaluated as easy. Install Claude Code, export your ANTHROPIC_API_KEY, run `/plugin marketplace add Imbad0202/academic-research-skills`, and run `/plugin install academic-research-skills`. No coding is required.
Alternatives and comparisons
Synthesizes topic outlines through retrieval and multi-perspective question asking to generate full-length articles with citations from scratch.
Provides a curated collection of n8n workflow templates for various AI agent use cases.
GitHub project description: Academic Research Skills for Claude Code: research → write → review → revise → finalize
README: ARS is built on the premise that **a human researcher augmented by AI avoids these failure modes better than either alone**.
README: This tool won't write your paper for you. It handles the grunt work — hunting down references, formatting citations, verifying data, checking logical consistency
README: ARS never runs experiments
README: - **Deep Research** — 13-agent research team with Socratic guided mode, PRISMA systematic review, intent detection, dialogue health monitoring, optional cross-model DA, Semantic S…