<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> This AI skill serves as a co-pilot for humanities and social science researchers, tackling thesis writing challenges like topic selection, literature hallucination, and text quality. It provides a comprehensive methodology, a deterministic text evaluation engine, and an extensive terminology database to ensure academic rigor and streamline the writing process from ideation to publication.

How It Works

The skill integrates a structured methodology guiding users through thesis development. Its core is a deterministic text evaluation engine with 21 rules (regex, term matching) for rigorous checks on credibility, terminology, logic, and structure, minimizing LLM self-assessment reliance. It automates English literature searches via free APIs (OpenAlex, Semantic Scholar, CORE, CrossRef) and guides users for Chinese literature, followed by AI analysis. Key features include full-process methodology, 350+ bilingual terms, and anti-hallucination guardrails.

Quick Start & Requirements

Installation options include direct copy-pasting of SKILL.md into AI platforms (Coze, Kimi, ChatGPT GPTs), installing as a skill in Claude Code/OpenClaw agents, or via command line: git clone, pip install -r requirements.txt, then running scripts/search.py or scripts/review.py. Requires Python 3.9+. English literature search uses free, stable APIs; Chinese literature search scripts are experimental. GitHub Repo.

Highlighted Details

• Full-Process Methodology: Guides topic selection, theoretical frameworks, material analysis, and revision diagnosis.
• Deterministic Text Evaluation: 21 rules across six dimensions (credibility, terminology, format, style, logic, structure) for rigorous quality checks.
• Extensive Bilingual Terminology: 350+ terms across 19 disciplines aid precise academic language.
• Automated English Literature Search: Leverages OpenAlex, Semantic Scholar, CORE, and CrossRef APIs.
• Anti-Hallucination Guardrails: Specific rules (R1-R4) prevent fabricated citations and data.

Maintenance & Community

Developed by @ganzhi-black. Encourages community contributions via Pull Requests for expanding the theorist database, terminology, evaluation rules, and data sources. No explicit community channels (Discord/Slack) are listed.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration into closed-source projects.

Limitations & Caveats

This tool is an assistant, not a ghostwriter; users must provide original arguments and manually verify all generated content, especially literature details. Automated Chinese literature search scripts are experimental and less reliable due to anti-scraping measures, recommending manual user searches instead.